In this book the authors explore current issues in the conceptualisation, assessment and treatment of antisocial personality. Consideration of two related constructs, psychopathy and sociopathy, is unavoidable given overlap in their meanings. The New Oxford Dictionary of English offers the following definition of ‘antisocial’: contrary to the laws and customs of society; causing annoyance and disapproval in others, e.g., aggressive and antisocial behaviour [1]. A person with antisocial personality is therefore one who both acts unlawfully or contrary to social norms and customs, and behaves in ways that others find objectionable. The dictionary defines a ‘sociopath’ as one with a personality disorder manifesting itself in extreme antisocial attitudes and behaviour.Footnote ¹ A ‘psychopath’ is defined as a person suffering from chronic mental disorder with abnormal or violent social behaviour. This latter definition perhaps reflects the common (layperson’s) perception of the ‘psychopath’ as someone who is dangerous and mentally deranged. Implicit in these definitions is the idea that there exists a continuum of antisocial behaviour or ‘antisociality’, ranging from the obnoxious but relatively benign, through the more severe and disordered (sociopathy) to the extreme (psychopathy) characterised by abnormal or violent social behaviour. This idea of an antisociality continuum is reflected in contemporary usage of these terms.

Lykken, a pioneer of contemporary psychopathy research, drew a distinction between the sociopathic individual or ‘sociopath’ and the ‘psychopath’ [Reference Lykken2]. While they were said to share a lack of the restraining influence of conscience and of empathic concern for other people, Lykken contended that sociopaths’ unsocialised character is due primarily to parental failures rather than inherent peculiarities of temperament; they are ‘the feral products of indifferent, incompetent or over-burdened parents’ [Reference Lykken2, p. viii]. In contrast, psychopaths’ inherent peculiarities of temperament make them unusually intractable to socialization. Although subsequent research has not supported Lykken’s distinction between these two types of antisociality in terms of their aetiology, subsequent research reviewed by Iacono has validated the existence of aetiologically distinct variants of antisociality – primary and secondary psychopathy – that are both appreciably heritable [Reference Iacono and Patrick3]. This distinction has stood the test of time and is considered in greater detail in Chapter 4. Iacono points out that parenting, especially from fathers, is important to the socialization of children, as we will see when, in Chapter 3, we consider family factors in the development of antisocial personality.

All three terms – antisocial, sociopathic, psychopathic – have appeared in psychiatric nosologies, the most important of which are the various iterations of American Psychiatric Association’s Diagnostic and Statistical Manual (DSM [4]) and the World Health Association’s International Classification of Diseases (ICD [5]). The latter has adopted the term ‘dissocial’ to mean much the same as ‘antisocial’. Below we will review how these various constructs have been treated in the various iterations of DSM and in ICD-11. Interesting to note at the outset is that ‘psychopathy’ and ‘antisocial personality disorder’ (ASPD) have been uneasy bedfellows, often diverging but at times merging together. Some psychiatrists (e.g., [Reference Mullen6]) have been downright antipathetic toward the construct of ‘psychopathy’. Conversely, some psychologists have been damning in their view of ASPD. Lykken, for example, stated: ‘Identifying someone as “having” ASPD is about as nonspecific and scientifically unhelpful as diagnosing a sick patient as having a fever, or an infectious or neurological disorder’ [Reference Lykken2, p. 23]. Lykken considered that ASPD comprises a family of disorders, the largest and most important of which is the ‘genus’ of sociopaths. We return to psychopathy throughout book and will examine in greater detail the ASPD construct. We will first briefly review the historical development of the ASPD construct in the various iterations of DSM (here we draw heavily on the review by Crego and Widiger [Reference Crego and Widiger7]).

ASPD in DSM

When the first edition of DSM (DSM-I [8]) appeared in 1952 it included a category termed ‘sociopathic personality disturbance’ encompassing a range of problems including sexual deviations, addictions, and a condition referred to as ‘sociopathic personality disturbance: antisocial reaction’ marked by persistent aggression and criminal deviance. When the first revision of DSM (DSM-II) appeared in 1968, the term ‘reaction’ was eliminated; sexual deviations, addictions and delinquent personality types were grouped together under ‘Personality Disorders and Other Non-Psychotic Mental Disorders’, which included a condition referred to as ‘antisocial personality’ [9]. Aligning the ASPD construct more closely with Cleckley’s psychopath prototype [Reference Cleckley.10], a person with antisocial personality was said to be grossly selfish, callous, irresponsible, impulsive, unable to feel guilt or to learn from experience and punishment, and to have a low tolerance of frustration. Individuals with antisocial personality were said to repeatedly come into conflict with society, to have a low frustration tolerance and a tendency to blame others for their problems. It further specified that a mere history of repeated legal or social offences was not sufficient to justify this diagnosis. The third edition of DSM [11] and its revision (DSM-III-R) saw much greater emphasis being placed on overt behaviour in its definition of ASPD. The nine items in DSM-III were childhood conduct disorder (required), along with poor work history, irresponsible parenting behaviour, unlawful behaviour, relationship infidelity or instability, aggressiveness, financial irresponsibility, no regard for the truth, and recklessness [12]. Explicit criterion sets were stipulated, with each criterion having relatively specific requirements. For example, recklessness required the presence of ‘driving while intoxicated or recurrent speeding’. The intention was to obtain greater diagnostic reliability, but in doing so, validity was sacrificed. From a psychological point of view this move toward behavioural criteria was a retrograde step, since it neglected the psychological factors such as motivational and emotional goals that may underlie these behaviours. Classifying people by their actions rather than by their psychological dispositions or traits may be suitable for purposes of criminal law, but it neglects the variety of reasons for any given action [Reference Lykken2]. Speeding can have a variety of motivations, for example, fear of missing an urgent appointment or a desire for the thrill of driving fast, and may be accompanied by quite different emotions (anxiety or exhilaration). New to the DSM-III-R criterion set was the lack of remorse, along with impulsivity or failure to plan ahead [12]. DSM-III-R shifted all of the personality disorders to polythetic criterion sets, requiring the presence of only a subset of features for an ASPD diagnosis.

The appearance of DSM-IV in 1994 marked a move away from DSM-III’s and DSM-III-R’s emphasis on behavioural criteria, many of which were removed [13]. It was the intention of the authors of the DSM-IV ASPD to shift the diagnosis closer to the conceptualization of psychopathy embodied in the Psychopathy Checklist (PCL) developed by Robert Hare in the early 1980s and revised (PCL-R; see Box 1.1) in 1991 [Reference Hare14]. This revision deleted two items from the original 22-item checklist (drug and alcohol abuse, and a prior diagnosis of psychopathy) and broadened the irresponsibility item to involve behaviours beyond simply parenting. The items of the PCL-R fall conceptually and statistically into distinguishable sets, or factors [Reference Hare, Neumann, Mokros and Patrick15]. Factor 1 (‘interpersonal/ affective’) comprises separate interpersonal and affective facets. Factor 2 (‘unstable and antisocial lifestyle’) comprises lifestyle and antisocial facets. Two items shown in Box 1.1, sexual promiscuity and having many short-term marital relationships, contribute to the total PCL-R score but not to any of the factors or facets.

The criteria for ASPD in DSM-IV (and retained in the main section of DSM-5 [16]) are shown in Box 1.2. At least three criteria are required for an ASPD diagnosis. An additional requirement is evidence of childhood conduct disorder (CD), but the number of CD criteria which had to be fulfilled was not specified. The chief difference between ASPD as defined in DSM and psychopathy as defined by PCL-R is inclusion in the latter of the interpersonal/affective features asterisked in Box 1.1. Hence ASPD is more closely related to PCL-R Factor 2 than to Factor 1. Psychopathy can be considered a more severe variant of antisociality than ASPD; thus, while about 75% of prison inmates were said to meet criteria for ASPD, only 15–25% were said to meet criteria for psychopathy [Reference Hare17]. These figures need to be qualified by recent results from a large database of US prison inmates [Reference Dargis18]. Of 1,000 prison inmates, 42% met criteria for ASPD, and of 4,600 inmates, 22% met the criterion for PCL-R psychopathy (total score ≥ 30). Compared with offenders with ASPD only, those whose ASPD co-occurs with psychopathy show more severe criminal behaviour [Reference Riser and Kosson19]. Those showing a triple comorbidity (ASPD comorbid with both psychopathy and borderline PD, characterized by a pervasive pattern of instability; see Box 1.3) were reported to show especially severe violence in their criminal history [Reference Howard, Khalifa and Duggan20]. Thus although psychopathy appears to lie toward the high antisocial end of the prosocial-antisocial continuum, it gives rise to a more severe manifestation of antisociality when combined with ASPD and borderline PD.

By the time DSM-5 appeared in 2013, there was considerably more research concerning psychopathy than ASPD. This was due to the large impetus to psychopathy research resulting from development of the PCL-R, which, while requiring special training, did not require expertise or training in psychiatry. It again appeared to be the intention of the DSM-5 work group to shift the diagnosis of ASPD toward PCL-R and/or Cleckley psychopathy. This was explicitly evident in the proposed (but subsequently rejected) change in name from ‘antisocial’ to ‘antisocial psychopathic’. Despite increasing indications that the features of PD are continually distributed and do not form discrete categories in nature, the APA Board of Trustees voted to retain the DSM-IV diagnostic system for personality disorders virtually unchanged in the main section of DSM-5. A brief description of each of these PDs is given in Box 1.3.

Among the many shortcomings acknowledged to attach to diagnostic categories of PD are their excessive comorbidity, their heterogeneity and their limited clinical utility. To this list may be added their impoverished and limited criteria. For example, hostility, sadism, lack of empathy, lack of insight, self-importance and power-seeking are arguably defining features of antisocial personality disorder, yet these aspects of mental life are absent from the DSM description. Contrast the truncated criteria for ASPD offered by DSM-IV/5 with the far more comprehensive description offered by Shedler and Westen [Reference Shedler and Westen21]:

Despite retaining the DSM-IV PD categories in the main section of DSM-5, it was decided to include an ‘alternative DSM-5 model for personality disorders’ in Section 3 of DSM-5, the section referred to as ‘Emerging Measures and Models’. This alternative, hybrid model will be described below.

If ASPD is defined, at least in part, in terms of doing things that could result in arrest (criterion 1 in Box 1.2), then naturally a large number of incarcerated persons will appear to suffer from the disorder. A sharper focus on the individual symptoms listed in Box 1.2 reveals particular problems with ASPD as a diagnostic category. Thus in a study by Schnittker and colleagues, the presence and symptoms of ASPD were explored among people with varying degrees of contact with the American criminal justice system (CJS) [Reference Schnittker, Larimore and Lee22]. Overall, nearly half of all respondents, who comprised 5,001 adults resident in the community, had some exposure to the CJS. Results indicated that contact with the CJS appeared to exert a disproportionate influence on an ASPD diagnosis. The prevalence of ASPD using the standard criteria (three or more of the symptoms listed in Box 1.2) was 14%. When symptom 1, failure to conform to social norms as indicated by having been arrested, was eliminated from the diagnosis, the prevalence of ASPD was reduced by more than 50%, even among formerly incarcerated persons. Some symptoms, in particular irritability/aggressiveness and irresponsibility, appeared to be linked to the presence and length of incarceration. This led the authors to suggest that the symptoms of those previously incarcerated might have been driven by their circumstance rather than by their personality. Last, and perhaps most important, criterion 7, lack of remorse, was met by only 5% of the overall sample and did not distinguish those receiving an ASPD diagnosis from those not receiving this diagnosis. Of the ASPD criteria listed in Box 1.2, only criteria 2 and 3 (deceitfulness and impulsivity) appeared to differentiate, to any substantial degree, those with from those without an ASPD diagnosis. We should note that assessment of ASPD in this study was carried out by lay interviewers rather than by mental health professionals. This might have led to the prevalence of an ASPD diagnosis in the sample being inflated. Despite its limitations, this study clearly indicates the need for a sharper focus on individual ASPD criteria so that criminality can be disaggregated from personality symptoms. Importantly, ASPD as specified in DSM does not appear to adequately capture the construct of insensitivity that, according to Tyrer [Reference Tyrer, Farnam, Zahmatkesh, Sanatinia, Black and Kolla23], is one of its key features (see Box 1.4 below).

Gender Differences in ASPD

ASPD is three times more prevalent in men than in women, less than 1% of whom are reported to receive this diagnosis [Reference Bishop, Völlm, Khalifa, Black and Kolla24]. We should note that many gender differences have been observed in ASPD with regard to its prevalence, risk factors, aetiology, genetic underpinnings, comorbid disorders, prognosis, key traits and symptoms, and its overall presentation (summarised in Table 1 in [Reference Bishop, Völlm, Khalifa, Black and Kolla24]). Males and females with ASPD differ in predisposing factors, offending behaviours, deceitfulness and impulsivity (more likely in females), aggression and recklessness (more likely in males), relationship problems (more promiscuity in males, greater marital separation in females), substance misuse (common in females, but highly prevalent in males), comorbid internalising disorders (more common in females) and narcissistic PD (more common in males). These authors point out that there is still a dearth of research carried out on women with ASPD, and very little is understood about why these gender differences occur. There is a significant proportion (somewhere between 30% and 50%) of female offenders with the disorder who would benefit from specific interventions and treatment programs, as well as customized assessment tools. We will consider gender differences in ASPD in greater detail when we examine its epidemiology in Chapter 5. In Chapter 3 we consider the possibility that developmental pathways to adult antisociality are gender-linked.

DSM-5 Alternative Model for Personality Disorders (AMPD)

In the Section 3 alternative model, the essential criteria to define any personality disorder are, first, moderate or greater impairment in personality functioning (criterion A) and, second, the presence of pathological personality traits (criterion B). As defined in this model, personality functioning consists of the degree to which there is an intact sense of self (involving a clear, coherent identity and effective self-directedness) and interpersonal functioning (reflecting a good capacity for empathy and for mature, mutually rewarding intimacy with others). This hybrid model requires assessment of the level of impairment in relation to six specific personality disorder types (antisocial, avoidant, borderline, narcissistic, obsessive-compulsive, and schizotypal) with the option for a diagnosis that is trait specified. Pathological personality traits are organized into five trait domains (negative affectivity, detachment, antagonism, disinhibition and psychoticism), each of which is further explicated by a set of trait facets reflecting aspects of the domain itself. A self-report instrument, the Personality Inventory for DSM-5 (PID-5), has been developed to measure these traits and their facets [Reference Krueger, Eaton and Clark25]. This trait system has been shown to correlate well with the Five-Factor Model, as shown in Figure 1.1. Watson and Clark [Reference Watson and Clark26] showed that the AMPD could be realigned to enhance its convergence with the Five-Factor Model of personality. A principal factor analysis based on these authors’ revised PID-5 yielded a clear and well-defined ‘Big Four’ structure comprising neuroticism/negative affectivity, antagonism (vs agreeableness), extraversion (vs detachment) and conscientiousness (vs disinhibition). Watson and colleagues have shown that some, especially agentic, aspects of extraversion have important links to personality pathology, for example, recklessness and exhibitionism [Reference Watson and Clark26, Reference Watson, Ellickson-Larew and Stanton27].

Figure 1.1 DSM-5 Section 3 domains and their Five-Factor Model counterparts.

From [Reference Miller, Sleep and Lynam31].

The trait facets that are diagnostic for an ASPD diagnosis (criterion B) in the alternative model, together with criteria for impairment (criterion A), are shown in Box 1.5. The trait domains and facets can be measured either using the PID-5 [Reference Krueger, Eaton and Clark25] or by clinician ratings, for example, using the DSM-5 Clinicians’ Personality Trait Rating Form (PTRF [28]). An interview-based instrument, the Semi-Structured Interview for Personality Functioning DSM-5 (STiP-5.1), has been developed to determine the severity of personality impairment [Reference Hutsebaut, Kamphuis, Feenstra, Weekers and de Saeger29]. Wygant and colleagues developed an interview-based instrument, the DSM-5 ASPD Impairment Criteria Interview, to assess impairments in identity, self-direction, empathy and intimacy in ASPD and psychopathy [Reference Wygant, Sellbom, Sleep, Wall, Applegate, Krueger and Patrick30]. However, results indicated that only the measure of self-direction significantly predicted ASPD. It therefore remains to be seen whether impairments in identity/self associated with ASPD can be reliably identified and measured. We return to this question in the final section of this chapter and in the following chapter.

Box 1.5

The AMPD includes a psychopathic features specifier for the diagnosis of ASPD associated with interpersonal/affective features of psychopathy. These features are indicated by low scores on anxiousness (from the negative affectivity domain) and withdrawal (from the detachment domain) together with a high score on attention seeking (from antagonism). There seems little doubt that this alternative model represents a much closer alignment of ASPD with psychopathy. However, there is limited representation of PCL-R psychopathy traits in the alternative DSM-5 Section 3 model [Reference Miller, Sleep and Lynam31]. For example, grandiosity was not included within the dimensional trait description of ASPD nor even within the eventually added psychopathy specifier. Wygant and colleagues [Reference Wygant, Sellbom, Sleep, Wall, Applegate, Krueger and Patrick30, Reference Wygant, Engle and Sellbom32] examined whether, in male and female offender samples, the DSM-5 alternative model of ASPD had moved closer to the traditional construct of psychopathy relative to the behaviourally oriented Section 2 model. While indicating a resounding affirmative response to this question, their results suggested that two additional trait facets, namely, grandiosity and restricted affectivity, might usefully be added to the trait facets listed in Box 1.5. Wygant and colleagues’ results indicated that the psychopathy specifier facets (low anxiousness, low withdrawal and high attention seeking) aligned more clearly with interpersonal/affective features of psychopathy than did the Section 2 model of ASPD. When assessed by a self-report measure, the Minnesota Multiphasic Personality Inventory (MMPI), ASPD was associated with a broad spectrum of maladaptive personality traits and overlapped considerably with other PDs [Reference Sellbom, Brown and Bagby33]. This points in the direction of a general severity dimension of PD, an aspect that is emphasised in the revised ICD assessment of PD which we consider below.

Zimmerman and colleagues [Reference Zimmermann, Kerber, Rek, Hopwood and Krueger34] highlight an important question with regard to the AMPD, namely, whether impairments in personality functioning (criterion A) and maladaptive personality traits (criterion B) provide distinct or overlapping information. Empirical findings reviewed by these authors indicated that measures of criterion A (including similar measures of personality functioning) and criterion B were highly correlated. A review by Widiger and colleagues [Reference Widiger, Bach, Chmielewski, Clark and DeYoung35] indicated considerable overlap between criterion A deficits and criterion B traits, and that criterion A may be largely redundant in the assessment of PD. However, other authors support the idea that criterion A (self/other deficits) adds importantly to the assessment of PD. Most clinicians would argue that intrapersonal and interpersonal problems are core features of personality pathology, with evidence suggesting that they co-exist and are reciprocally interrelated in PD patients [Reference Beeney, Lazarus, Hallquist, Stepp, Wright, Scott, Giertych and Pilkonis36]. It can be argued that the way self/other deficits are formulated in AMPD criterion A does not adequately capture all there is to know about the individual’s sense of self and interpersonal relatedness or offer ideas on how to improve self and relational functioning. Two separate questions arise in relation to ASPD. First, do the self and interpersonal descriptions contained in DSM-5 criterion A adequately capture the intrapsychic and interpersonal deficits shown by individuals with ASPD? Second, does a description of ASPD in terms of the trait domains and facets as captured by AMPD measures (see Box 1.5) adequately cover core features of their interpersonal deficits? In the chapter that follows we will examine in greater detail the question of what exactly are the interpersonal deficits of antisocial individuals.

ICD-11

Compared with DSM-5, ICD-11 represents a more radical departure from the categorical system of assessing PD. ICD-11 has jettisoned all PD categories in favour of assessing the level of PD severity, ranging from mild to severe, with each level of severity qualified by five trait domains (negative affectivity, disinhibition, dissociality, anankastia and detachment). One exception to ICD-11’s eschewal of PD categories has been its retention of a ‘borderline pattern specifier’. The definition of PD in ICD-11 is shown in Box 1.6. DSM-5 Section 3 and ICD-11 share a twofold conceptualization of severity and style, but there are noteworthy differences. First, ICD-11 does not include the possibility to assign specific PD diagnoses (except borderline PD). Second, in ICD-11 the assessment of trait domains is not a necessary part of the diagnosis; for a diagnosis of PD in AMPD, at least one maladaptive personality trait domain or facet must be in the clinically significant range. Third, the trait domain of psychoticism and its trait facets are absent from ICD-11.

Both DSM-5 AMPD and ICD-11 emphasize, in broadly similar ways, severity as an important factor in assessment of PD. While DSM-5 and ICD-11 both emphasise interpersonal dysfunction as critical to a diagnosis of PD in general, ICD-11 gives this rather more emphasis than does DSM-5. For ICD-11, the severity of interpersonal problems is a key factor in defining overall severity of PD. A classification as mildly severe requires that there should be notable problems in many interpersonal relationships. For a classification as moderately severe there should be marked problems in most interpersonal relationships. A classification as severe requires that there should be severe problems in interpersonal functioning affecting all areas of life.

It should be noted, however, that other features are important in defining severity of personality dysfunction in ICD-11, namely: degree and pervasiveness of disturbances in functioning of aspects of the self; pervasiveness, severity and chronicity of emotional, cognitive and behavioural manifestations of the personality dysfunction; and, most importantly, risk of harm to self or others. A questionnaire developed to measure severity according to ICD-11, the Standardized Assessment of Severity of Personality Disorder (SASPD), operationalises PD severity largely in terms of interpersonal dysfunction; for example, it includes the items ‘being with others’, ‘trusting other people’ and ‘friendships’ [Reference Bach and Anderson37].

The dissociality trait domain in ICD-11 bears a close resemblance to the antagonism domain in the DSM-5 alternative model, having at its core ‘disregard for social obligations and conventions and the rights and feelings of others. The traits of callousness, lack of empathy, hostility and aggression, ruthlessness, and inability or unwillingness to maintain prosocial behaviour are characteristically present but not always displayed at all times’ [Reference Tyrer, Reed and Crawford38]. As noted above, Tyrer suggested that ASPD is best characterized by the Three I’s shown in Box 1.4 [23]. Tyrer and colleagues noted that ‘the presence of insensitivity … is perhaps the strongest component of psychopathy’ [Reference Tyrer, Reed and Crawford38].

What Motivates Engagement with Others?

It is a characteristic of dissocial individuals, as noted in the quotation above, that they are unable or unwilling to maintain prosocial behaviour. Here we briefly consider the question of what motivates most people to maintain prosocial behaviour and to act prosocially. Clearly, whatever this is, it is lacking in people with antisocial traits. This question has been recently addressed in a series of studies carried out by Lockwood and colleagues [Reference Lockwood, Hamonet, Zhang, Ratnavel, Salmony, Husain and Apps39, Reference Lockwood, Ang, Husain and Crockett40, Reference Contreras-Huerta, Lockwood, Bird, Apps and Crockett41]. Their results suggested that prosocial behaviour includes motivational and moral components, which are modulated by affective sensitivity. Their key finding was that people who showed a configuration of traits characterised by high empathy and affective reactivity were averse to harming others and were more willing to exert effort to benefit others. Lockwood and colleagues spoke of a deep prosocial disposition or affective sensitivity shown by people who engaged affectively with themselves and others; they were more likely to assume costs to benefit other people [Reference Lockwood, Ang, Husain and Crockett40]. These authors found that high empathic concern and low emotional apathy were the strongest predictors among several affective and psychiatric traits putatively suggested to relate to prosocial behaviour; they suggested that a lack of affective sensitivity may be associated with poor social relationships seen in a high number of mental health conditions [Reference Contreras-Huerta, Lockwood, Bird, Apps and Crockett41]. They further suggested that in order to recognize and care about others’ emotions and act accordingly, people may have to be sensitive to the experience of emotions and be motivated by them. While most people are motivated to engage emotionally with others – presumably because such emotional engagement is, for most people, pleasant and rewarding – antisocial individuals can be presumed to lack this motivation, showing what Lockwood and colleagues referred to as ‘prosocial apathy’. A similar view is expressed in a recent reformulation of psychopathy that viewed it as stemming from strategic, motivated processes [Reference Groat and Shane42]. This reformulation sees psychopathic individuals’ lack of empathy and concern for others as reflecting a lack of motivation to care for others rather than an inability to care. The reason they are not motivated to engage empathically with others is that they do not place value on the sharing of emotions.

Recent findings from Lockwood and colleagues [Reference Cutler, Wittmann, Abdurahman, Hargitai, Drew, Husain and Lockwood43] suggest that despite declines in learning ability associated with ageing, motivation could play a role in preserving learning to help others (‘prosocial learning’). These authors found an inverse relationship between psychopathy and prosocial learning, but only for older adults. This suggests that age-related differences in prosocial learning could be linked to basic shifts in individual traits and motivations over the lifespan. We will return to this issue of what motivates dissocial/psychopathic individuals at various points throughout this book.

Summary: DSM-5 and ICD-11

We can summarise the foregoing by saying that three main themes have emerged in both DSM-5 and ICD-11 with respect to how PDs are conceptualised and assessed. First, there has been an increasing focus on PD as comprising continuous rather than categorical variables. Second, there is an increasing recognition that PD varies critically along a severity dimension, from mild to severe. Third, there has been an emerging consensus that interpersonal dysfunction lies at the heart of PD. Despite these changes, there is recognition that the ways in which PD is currently conceptualised and assessed remain unsatisfactory. We turn now to examine recent critiques of DSM-5 and ICD-11, and ask the question: After ICD-11 and DSM-5, where next? In the final chapter we suggest a possible way forward.

Critiques of DSM-5 and ICD-11

Detailed critiques of the current state of play regarding the conceptualisation and assessment of PD have been offered by Clark and colleagues [Reference Clark, Vanderbleek, Shapiro, Nuzum, Allen and Daly44], Livesley [Reference Livesley45] and Huprich [Reference Huprich46]. Clark and colleagues highlighted inadequacies of the hybrid (type/trait) approach adopted in the DSM-5 alternative model. They argued that a severity diagnosis (criterion A) plus a full trait-specified description was sufficient and therefore a categorical diagnosis was superfluous. They supported this conclusion by showing that the overwhelming majority of people with pathological personalities do not fit prototypical PD-type descriptions and that their personalities are typically more complex than a categorical PD system would imply, even when they meet diagnostic criteria for a single PD. Diagnosis by PD type implies homogeneity where none, in fact, exists. This was exemplified in the case of two patients who both met categorical criteria for antisocial, borderline and narcissistic PDs but one of whom had a far broader range of maladaptive traits than the other. Attaching diagnostic labels to these patients is misleading since to do so would falsely imply that the trait constellation is the same in both cases. This recognition of what Wright and Woods have referred to as the ‘staggering heterogeneity in how each individual functions’ [Reference Wright and Woods47] has been highlighted by those who advocate a personalised (idiographic) approach to psychiatric diagnosis. This shifts the diagnostic focus to dynamic processes within individuals, with no assumption that these processes are homogeneous, even among individuals with the same diagnosis. While in its infancy, this personalised approach is likely to prove increasingly influential and is one that the current authors favour for the assessment of antisocial personality, as outlined in the final chapter.

Both Livesley [Reference Livesley45] and Huprich [Reference Huprich46] advocate a thoroughgoing re-evaluation of the PD construct. Both want to move the field beyond both traits and categories of PD in order to mitigate the shortcoming seen in current diagnostic systems. Both advocate a reconceptualization of PD that includes a focus on phenomenological aspects of personality pathology. Most importantly, we are urged to ‘think more carefully about what personality is, and how it becomes pathological … It is time to evolve to the next level’ [Reference Huprich46, pp. 688–689]. Livesley too argues that a new course needs to be charted, one that, first, incorporates concepts from normal personality science and, second, includes multiple levels of explanation [Reference Livesley45]. The latter fall into two groups: those concerned with publicly observable phenomena (neurobiology and observable behaviour) and those concerned with private mental events – cognitions, intentions, meaning systems.

Livesley [Reference Livesley45] identifies several impediments to progress in the PD field. First, current conceptualisations of PDs – whether viewed as discrete categories or as constellations of traits – fail to do justice to the phenomenological and aetiological complexity of personality disorder.

A second impediment to progress is that current conceptualisations of PD adopt a version of the medical model and, relatedly, an essentialist philosophy – the idea that disorders have an underlying nature or pathology. For Livesley, the essentialist philosophy espoused by DSM-5 impedes a rational classification of personality disorders by encouraging ‘an endless search for real types and encouraging conservative revisions by implying that when diagnostic criteria fail to function as designed, the appropriate response is to revise them rather than re-evaluate the diagnostic construct itself’ [Reference Livesley45]. Livesley’s criticism was echoed by Wilshire and colleagues, who emphasised that the categorical approach to psychopathology espoused by DSM-5 rests on the dubious assumption that its symptom profiles are meaningful indicators of the nature of some underlying disorder [Reference Wilshire, Ward and Clack48]. This carries with it the further implication that we are ultimately seeking explanations phrased at the biological level of description; but Wilshire and colleagues point out that ‘at least some aspects of DSM-5 disorders may be better captured at a psychological or even a social/cultural level of description’ [48, p. 9]. The general thrust of the remainder of this book is that this is especially true of ASPD.

The third impediment to progress is the lack of an adequate alternative model. Livesley rejects trait-based models since traits are largely descriptive, atheoretical, and non-explanatory. Traits are not sufficient to capture individual difference dimensions adequately for clinical purposes since they focus primarily on structural and static features and neglect the personality processes crucial for understanding and treating PD. In short, trait models have just too many problems to provide the foundation for an official classification of PD.

A fourth and final impediment to progress is represented by extraneous (non-scientific, political) influences on the development of PD classification systems. Those tasked with revising PD classification systems were themselves heavily invested in research programmes based on the very diagnostic concepts that needed extensive revision or even outright elimination.

Livesley asserts that we need to consider what clinicians need to treat patients and what investigators need to investigate PD. Some authors regard the clinical goal of classification – to aid practitioners in identifying, conceptualising and treating conditions – as increasingly separate from a productive research classification [Reference Blashfield, Keeley, Flanagan and Miles49]. Huprich points out that a description of a patient’s traits may not be relevant to their therapy, which is designed not to modify trait domains or their facets but to reduce the problems that arise from internal or interpersonal distress. Moreover, a purely trait-based system may fail to offer the degree of differentiation needed to separate a personality disorder from other conditions, or indeed to differentiate qualitatively different types of personality pathology. As configured in the alternative model, ASPD shares many traits with other PDs, yet there is reason to think there may be core traits, for example, insensitivity, lack of empathy, and callousness (see Box 1.4), that differentiate it from other PDs. Huprich questions whether or not PD categories should be retained, and gives qualified support for their retention. He argues convincingly that the use of categorical and prototypical thinking is inevitable even within a dimensional system, for example, in the decision to classify someone with a PD as ‘severe’ or ‘not severe’ in the ICD-11 classification system. Huprich maintains that certain categories, in particular, borderline PD and ASPD, remain clinically useful and an important part of the diagnostic nomenclature.

Livesley asserts that what the clinician primarily needs to know is, first, whether a patient has PD and, second, how severe it is. He proposes a multifaceted framework that can be flexibly tailored to meet diverse clinical or research needs. Specifically, he proposes three facets. (1) The first is a diagnostic facet consisting of a definition of general personality disorder and a way to evaluate severity. Key elements are the degree of impairment to self and interpersonal functioning. (2) The second facet is a structured assessment facet consisting of clinically important personality constellations and a lexicon of specific traits and related structures and mechanisms. To capture the complexity and nuances of PD a diagnostic assessment needs to embed diagnostic features within a narrative that captures the richness of clinical presentations, including more inferential features such as emotions, schemas, meaning systems, traits, relationships, conflicts, and self-problems. (3) The third facet is a functional impairment facet consisting of four broad domains of personality dysfunction: symptoms (e.g., emotional intensity and reactivity, deliberate self-injury, dissociative reactions, quasi-psychotic symptoms and regressive behaviour); regulatory and modulatory mechanisms (emotional regulation capacity, impulse control, executive functions, metacognitive functioning and capacity for self-reflection); interpersonal problems (interpersonal schemas and patterns, conflicted relationships, interpersonal boundaries, and the capacity for intimacy, cooperation, empathy and altruism); and self-pathology (core self-schemas, experienced authenticity of self-states, stability of self-states and self-representations, and self-narrative).

A Role for Motivation in Personality Pathology

Huprich’s above-mentioned challenge divides into two questions: First, what is personality? Second, how does it become pathological? In the following chapters we argue that motivation is intrinsically linked to personality – indeed, within mainstream personality psychology there is an increasing integration of motivation and personality [Reference Zeigler-Hill, Vrabel, Sauls and Lehtman50]. The integration of motivational processes and motivation-related constructs into current conceptualizations of personality offers the possibility of a deeper understanding of both motivation and other components of personality (e.g., personality traits) [Reference Zeigler-Hill, Vrabel, Sauls and Lehtman50]. If motivation is an intrinsic part of normal personality, it follows logically that it must also be an intrinsic part of personality pathology. Studies (e.g., [Reference Zeigler-Hill and Hobbs51]) have started to investigate the relationship between maladaptive personality traits, as operationalized using the DSM alternative model, and fundamental social motives. This framework assumes that different relationships are characterized by unique adaptive problems that must be managed in specific ways. Findings suggested that the individuals who scored high on the antagonism trait, which – as we shall see in the next chapter – captures a malign and distrustful interpersonal style, were motivated to seek status via dominance strategies (e.g., ‘I am willing to use aggressive tactics to get my way’) but were poorly motivated to protect self or others [Reference Zeigler-Hill and Hobbs51].

If we want to understand what people are like and why they behave the way they do, we need to look at both traits and values – the latter being goals that people find desirable and use as guides for their behaviour across different situations [Reference Fischer52]. The use of network analysis (an alternative to factor analysis) to interrogate individual differences data has afforded new insights into how human values, together with neurobiologically derived motivations and personality traits, form a complex network structure comprising three basic dimensions of motivation: behavioural approach versus behavioural inhibition, exploration versus constraint, and self- or ego-orientation versus social orientation [Reference Fischer and Karl53]. This last dimension, reflecting a combination of extraversion and agreeableness (the inverse of antagonism), clearly reflects differences between individuals in their propensity to act in an antisocial way – to contravene social norms and values.

Among the many symptoms in the Shedler–Westen Assessment Procedure (SWAP-200) rated by psychiatrists and clinical psychologists as prototypical of PD cases [Reference Shedler and Westen54], one can discern two core interpersonal deficits: (1) a readiness to interact with others but in maladaptive, self-defeating or antisocial ways and (2) a disengagement from interaction with others, motivated by fear of the consequences of social engagement (see Figure 1.2). The PDs listed on the left side of Figure 1.2 all show a disposition to engage interpersonally, but in self-defeating and antisocial ways, while the PDs listed on the right are disposed to disengage interpersonally in maladaptive ways. This schema essentially re-describes PD in terms of an ‘approach versus withdrawal’ dimension that is fundamental to human motivation, and specifies particular motivations driving ‘approach’ and ‘withdrawal’, that is, specific fears and specific approach-based motives. It has the advantage that it knits together into a single schema the two aspects of impaired interpersonal functioning, namely, impaired capacity for intimacy and impaired socialization, and relates these to particular categories of PD.

Figure 1.2 Motives for interpersonal engagement versus disengagement associated with different personality disorders. Interpersonal symptoms of PD from SWAP-200 were identified by the current authors from those deemed prototypical of actual PD cases by a national sample of psychiatrists and clinical psychologists (54).

Recalling how PDs are arranged in the three clusters shown in Box 1.3, it will be noted that Cluster B PDs align in the left-hand column of Figure 1.2, while Cluster A and C PDs align in the right-hand column. Some PDs (borderline, histrionic, narcissistic) appear in both columns, indicating that they can be associated, at different times, with both maladaptive engagement and maladaptive disengagement. Narcissism, for example, bifurcates into two related but distinct dimensions, narcissistic grandiosity (marked by boldness and approach) and vulnerability (marked by reactivity and aversion) [Reference Krizan and Herlache55].

It is apparent from inspection of Figure 1.2 that the PDs listed in the left-hand column are, broadly speaking, externalizing disorders, while the PDs listed in the right-hand column are internalizing disorders. It was recently suggested that externalizing pathology might involve inappropriate wanting – and acquisition – of rewards (vs sharing resources fairly), whereas internalizing psychopathology might involve insufficient enjoyment of experiences that most people find pleasurable [Reference Watson and Clark.56]. Thus what all the PDs shown in the left-hand column of Figure 1.2 might have in common is their inappropriate and antisocial desires, for example, a desire to gratify oneself at another’s expense (ASPD), or to dominate another (narcissistic PD). We speculate that this may be driven by an over-active but dysfunctional approach motivational system and an over-active dominance motivation system [Reference Johnson, Leedom and Muhtadie57].

It is perhaps unsurprising that borderline PD appears in both columns of Figure 1.2, since it combines elements of both externalizing and internalizing [Reference Eaton, Krueger, Keyes, Skodol, Markon, Grant and Hasin58]. As a complex form of personality pathology, borderline PD putatively combines pathological desires with a pathological inability to derive pleasure (anhedonia). Several lines of evidence are consistent with the idea that anhedonia underlies the maladaptive disengagement shown by all the PDs shown in the right-hand column of Figure 1.2. First, anhedonia is an important and often overlooked symptom of borderline PD that contributes significantly to its severity [Reference Marissen, Arnold and Franken59]. Second, individuals with predominantly borderline features apparently lack the motivation to engage with their environment, including their social environment. For example, when experiencing a lack of positive affect (when feeling anhedonic) they experience a lack of ‘willpower’ [Reference Heiland and Veilleux60]. Third, the relationship between depression and aggressive and antisocial acts (lying, stealing and violating the rights of others) is explained by individual differences in anhedonia, putatively associated with weaker inhibitory control over aggressive and antisocial impulses [Reference Salem, Winer, Jordan and Nadorff61].

How Does Personality Become Pathological?

An answer to the second question raised by Huprich – how does personality become pathological? – requires a lifespan developmental approach to personality pathology and – in the current context – to antisocial personality. Other than perhaps in their early/mid-adolescent years, most people behave more or less prosocially. They are prepared, at least to a degree, to sacrifice their own self-interest in the service of acting in such a way as to benefit others. This is commensurate with Homo sapiens being an essentially prosocial (or ‘ultrasocial’) species whose mental apparatus evolved in our remote ancestors as an adaptation to the pressures of living in social groups. An evolutionary perspective suggests that the human brain is essentially a motivational device that has been shaped by evolutionary processes to promote adaptive behavioural responses to the sorts of recurring opportunities and challenges that humans have faced throughout the course of evolutionary history [Reference Zeigler-Hill, Vrabel, Sauls and Lehtman50]. Yet some individuals, a minority, emerge from their teenage years as antisocial – they show a lifelong disregard for social norms and conventions, often break the law, show an aggressive and intrusive interpersonal style and sometimes show violence toward others. They often show the hallmark features of ASPD – the Three I’s shown in Box 1.5. We need to know what goes wrong, when and how; and our ‘how’ explanation needs to include the different levels of explanation referred to above, none of which, as Livesley points out, is more important or fundamental than the rest.

Finally, problems of ‘selfhood’, as we have seen above, are regarded as an intrinsic part of personality pathology and are thought, in some way, to be linked to the interpersonal problems that are so characteristic of people who are classed as personality disordered. Regarding self/other functioning, Huprich suggests that the way self/other functioning is represented in DSM-5 and ICD-11 ‘may not adequately capture all there is to know about the individual’s sense of self and interpersonal relatedness or offer ideas on how to improve self and relational functioning’ [Reference Huprich46]. Current conceptualisations of PD lack a well-worked-out theory of self, and an explanation of how self-dysfunction (however defined) is related to interpersonal dysfunction. A coherent, psychologically informed theory about what constitutes ‘self’ has until recently not been available. The self is first and foremost an inherent duality of I and Me, as was acknowledged years ago by the American psychologist William James. We will argue with McAdams [Reference McAdams62] that psychologically speaking, the I/Me dynamic plays out in three different guises: the self as (1) social actor, (2) motivated agent and (3) autobiographical author. Stated succinctly, to know the self fully is to know the traits, the goals and the stories. We will therefore attempt to examine the question of what aspect(s) of self might be abnormal in relation to antisocial personality. This will necessarily be speculative since there is a lack of empirical data addressing this aspect of functioning in people with ASPD.

The chapters that follow are premised on the idea that PDs, like all mental disorders, are complex, multi-factorial problems which can be understood and approached from a variety of perspectives. We offer three complementary perspectives of antisocial personality: an interpersonal perspective, a developmental perspective and a neurobehavioural perspective. We then consider antisocial personality in clinical contexts, particularly in regard to its treatment, and examine ethical and legal issues arising from the treatment of people with ASPD. As will become clear as we proceed, we regard antisociality as being a continuum of severity, running from an ‘extremely prosocial’ pole at one end to an ‘extremely antisocial’ pole at the opposite end. Not only do individuals differ in regard to their position on this continuum; they may also move along it in the course of their lives. For example, most people move toward the antisocial pole as they enter adolescence but soon return to a prosocial position. At the extreme antisocial end of the continuum are to be found individuals whose antisociality is both serious and persistent throughout the lifespan, often involving interpersonal violence.