A taxonomy of personality traits in children and adolescents.
With this theoretical review, we intend to understand the relationship between adolescent psychopathy and adult psychopathy, taking into account three fundamental questions: (1) the conceptualization of personality, in other words, knowing the extent to which it makes sense to speak about a structured and defined personality in adolescents; (2) the notion of disorder quite present in the stage of adolescent development; (3) and finally, whether the previously alleged disorders are stable identities or if they have temporal continuity, extending into adult life. We are aware that there is no unanimity or consensus regarding the essential core of psychopathy, that is, it presents itself in distinct ways that make it difficult to configure it in taxonomical or dimensional terms of the personality, ignoring much about its etiology, these uncertainties, end up to be translated into a major openness to the study of pre‐adult population. Thus, the study of adolescent and juvenile psychopathy will depend a lot on how much we know about adult psychopathy. This does not imply that we do not carry out studies of a longitudinal and measuring nature through psychometric instruments that allow us to clarify the way this syndrome manifests itself and develops throughout life, taking into account the Big Five.
In a study about psychopathy in adolescence, it is inevitable to address the question whether there are, in this age group, individuals with this disorder or, at least to speak more prudently, whether we can safely identify, adolescent individuals with psychopathic traits. This problem was unnoticed for many years because the investigations, up to recent date, were initially oriented toward those who manifested expressed behaviors of psychopathy, that is, repeat offenders and highly violent adults, revealing high levels of callous‐unemotional traits. Among these investigations, only sporadically, there are descriptions of young psychopaths, as in Bowlby’s study  on juvenile offenders, but who would not at the time have followers.
In fact, the issue of juvenile and adolescent psychopathy only began to be seriously discussed just over 20 years ago, especially since the investigations of Frick et al.  and Lynam , coincident with the time when courts began to pronounce sentences on defendants from these age groups on the basis of psychopathy diagnoses (mostly in Canada but also in the US; see Ref. Frick ). This controversy emerged especially in the early twenty‐first century, with the publishing of various numbers of prestigious journals entirely dedicated to this subject (e.g., Law and Human Behavior, 26 (2), 2002 and Behavioral Sciences and the Law, 21, 2003), considering divergent views, with some authors being skeptical against the existence of this personality disorder (PD) before adult life as opposed to those defending the opposite view. The most skeptical  argued that many of the typical psychopathic traits were specific of their own or inherent to development in adolescence, and therefore, it was questionable that true psychopaths could be identified in this population stratum, or, in an even more radical way, it was assured that there are no juvenile psychopaths and adolescents, because, in a general way, personality has not fully stabilized until this age, and therefore, we cannot speak about personality disorders before adult life .
Basically, this emphasized controversy has been based on three major and compelling questions for those who propose to investigate psychopathy before adulthood, without having the ambition to bring things to a closure, namely: The first question relates to the conceptualization of personality itself. It is questioned in the debate the legitimacy to speak about the personality of an adolescent, as it would correspond to something that would only have its stable structure with the emergence of adulthood. The second question relates to the notion of disorder, or stating Wallon , it means knowing whether the adolescent disorders are “turbulences” inherent to their development, or if they are they likely to have a clinical‐forensic status. Finally, the third question relates to whether the alleged disorders are stable entities and if they have temporal continuity, extending into adult life. In the following chapters, we will answer these three questions.
1.1. The personality in childhood and adolescence—continuity and development
In popular psychology (folk psychology), in Bruner’s interpretation , personality exists before the age of 18. The parents of the children routinely refer to their children’s personality, whether they use that word, or use another word that is equivalents in the current language, for example, “temperament” or “way of being.”
The accumulated evidence in the last 10 years about the structure and development of juvenile and adolescent personality has shown that it has very similar characteristics to adult personality. The temperament traits organize themselves in a similar way to adult personality traits, as it has been repeatedly highlighted by Caspi and his co‐workers [9, 10]. Therefore, we can speak of a temperament organization, whose analysis was started by Thomas and Chess, in 1963, within a longitudinal study on temperament traits stability in babies and children, the New York Longitudinal Study . The original model of these authors distinguished nine temperament traits, which would be present since early childhood. More recent models, resulting from questionnaires and observation protocols, achieved, by factorial analysis, less temperamental traits, generally 6 or 7. Thus, Caspi et al.  list the following six traits, as being typical of contemporary research: level of activity; positive emotions/pleasure (distress); irritable/anger/frustration (distress); fearful/escape from new situations (including social situations); tranquility (tendency to remain calm) and ability to concentrate/persist. It is possible to match, at least partially, these six temperamental traits with the higher‐order personality traits as a result from adult research, the Big Five: Extraversion, Neuroticism, Agreeableness, Conscientiousness and Openness to Experience . For example, several factorial analyzes of questionnaires, adjective lists, and California Child Q‐Set produced factors similar to the Big Five and some of their subfactors (so‐called second‐order factors), in children and adolescents (see Ref. [10, p. 307]). As a result of these studies, Caspi and Shiner  elaborated a proposal for a taxonomy of first‐ and second‐order personality traits of children and adolescents, which will represent the personality of these age groups and is, as can be seen in Table 1, structurally identical to that of adults. This does not mean that there are no differences among children, adolescents, and adults as regards the delicate organization (i.e., in terms of the second‐order traits) of the personality. It is certain that some temperamental characteristics of children do not find accurate match in the next age stages (e.g., it is the case of irritability), and it is seen equally that the factorial composition of the Big Five is, in some cases, defined by different items in the instruments used in children and adults . However, in general, we can draw from these results the conclusion that it is not only legitimate to speak of personality before adulthood [10, p. 307], since the age of 3. We may or may not designate this personality by the word temperament, but it is also appropriate to say that in structural terms, the constituent elements of the adult personality have been available since childhood. But it cannot be inferred from it that the organization of personality is unchangeable. As Caspi et al. said, “although children exhibit traits that are remarkably similar to those found in adults, researchers should be aware of developmental differences in the manifestation of these traits; for example, the traits may be less coherent in early infancy” [9, p. 456].
|First‐order traits||Extraversion (E)||Neuroticism (N)||Conscientiousness (C)||Agreeableness (A)||Openness to experience (O)|
|Energy/Level of activity||Anxiety||Self‐control||Tendencies||Creativity|
|Low E + N||N + Low A||C + A|
The empirical analysis of stability versus life‐long personality change can be made in the perspective of Caspi et al. using three types of studies, called differential measurement, mean‐level measurement and ipsative measurement . Although different, these three types of study provide results that are complementary and whose integration is essential in order to have a complete and rigorous representation of the degree of personality stability.
Differential studies, of a longitudinal nature, compare, for each temperament or personality trait, and on several successive moments, the related positions of a group of individuals, using statistical procedures of test‐retest (rank‐order correlation). The objective is to know, for each measured trait (e.g., anxiety in children and young adolescents, or its equivalent in older adolescents and young adults), if these individuals maintain, in those successive moments, the same positions. As an example, if, in a group of 100 individuals, the result of a three‐year‐old child in a measure of sociability reveals her as the most sociable, and then at 12 years of age, its result will maintain her in the same position, we will say that there is differential stability, regardless of whether this result may vary in absolute value.
The so‐called middle‐level studies, meanwhile, aim to measure the average absolute value (mean‐level) of one or several first‐ and second‐order traits in a given population, in order to determine the extent to which the different aspects of personality retain, throughout life, and especially in moments considered as transition, the same average operative expressiveness. In other words, we want to determine to what extent the overall structure of personality, expressed in terms of the absolute values of its different components, varies over time for the population as a whole. While it is desirable that these studies be longitudinal, thus ensuring that temporal fluctuations in trait intensity are measured in the same individuals, there are other investigations that use a cross‐sectional methodology, measuring at the same time the mean level of traits in different cohorts.
Both the differential and the mean‐level investigations have as a common element the fact of comparing a group of individuals regarding the results (relative and absolute) obtained in one or several components of the personality. That is, there are studies that seek to determine continuity (stability), in this case, the statistical distribution of the results of a group or, inferentially, the population of individuals in these traits (the variables), ignoring their specific configuration for each individual. They are, therefore, investigations focused on the variables. On the other hand, ipsative longitudinal studies in Block’s (1971) are people‐centered as they seek to determine how, in each person, the personality structure—defined by the results of each trait or component—varies over time. For this purpose, this internal structure is measured in two or more moments of time by Q‐Sort type methodology, obtaining an overall coherence index (by Q‐Sort correlation) that represents the degree to which the individuals of a sample will vary over time.
After analyzing the methodologies, we will present the main empirical conclusions pertaining to these three types of studies. To that end, we will rely mainly on some recent meta‐analyzes performed by Roberts and DelVecchio (2000), and by Roberts, Walton, and Viechtbauer (2006), as well as two excellent syntheses produced by Caspi and his collaborators (Caspi et al., 2005; Caspi & Shiner, 2006).
1.2. Differential Continuity (rank‐order) of personality
The meta‐analysis that Roberts and Delvecchio (2000)  carried out, taking as reference organization the Big Five, considered 152 differential personality studies, allowing the following conclusion: (a) from childhood to adulthood, test‐retest correlations (rank‐order) are moderate, with an average value after correction for the reduction of 0.40; (b) stability tends to increase with age, from 0.41 in childhood to 0, 55 at age 30, reaching a maximum value of 0.74 between ages 50 and 70 (reviewed values); (c) the stability decreases as measurement is more spaced in time; (d) it does not vary between traits; (e) neither with the assessment method; and (f) nor by gender. These results show, according to Caspi et al. [9, p. 466–467], that the magnitude of personality stability is overall impressive and only exceeded, in all psychology, by the stability of cognitive ability measures and interests referring specially to adolescence, as well as the fact, “that the level of stability increases in an approximately linear way through adolescence and first stage of young adulthood.”
1.3. Stability and change in mean‐level
In an investigation led by Roberts et al. , 92 studies were reviewed on stability and change of mean‐level of personality domains, organized according to Big Five and its features, throughout life. The results of that meta‐analysis showed that: (a) pertaining to extraversion, the side of social dominance (assertiveness, dominance) increased since adolescence until middle age, especially in the first stage of adulthood (20–40 years of age), whereas the side of social vitality (sociability, talkativeness) increased in adolescence and decreased in adulthood; (b) in agreeableness and conscientiousness, the results increased in the first stage of adulthood and middle age; (c) the neuroticism traits decreased equally during adult life (20 years upwards); and (d) openness to experience traits increased in adolescence and in the first stage of adulthood, and decreased in old age.
A surprising result shows that the most mean‐level personality traits change occurs during the first stage of adult life between 20 and 40, Caspi and Shiner [10, p. 337], suggested that the investigation of the personality maturation mechanisms should be focused at this age.
1.4. Ipsative stability and change
There are very few longitudinal studies that have used Block’s (1971) Q‐Sort Method to measure the internal change in personality structure of individuals. This is disappointing, because of the above‐mentioned differential and mid‐continuity research, concentrating on the overall variance of traits in groups or populations, tend to encapsulate the cases of individual variation that occur in them. That is to say, no one tells us what percentage of individuals in these groups exhibits internal structural stability, and how many are those with moderate or marked variations. It is not clear, how many people in each cohort are stable and how many are not. In Block’s (1971) study, this methodology was used in a pioneering way, with great differences in the degree of personality stability among individuals. Although the Q‐Sort correlation showed high stability for the groups as a whole—0.70 between childhood and the end of adolescence and 0.50 between this and adult life—these results concealed the enormous variability between individuals. For example, the intraindividual Q‐Sort correlation, that is, for each individual, varied between median negative values and positive values so high that they were only limited by the measurement error margin. This means that in each cohort, there are people whose trait structure, measured by the Q‐Sort correlation over time, is invariant, and others in which the personality seems to undergo severe structural inversions. Similar results were obtained in more recent studies, with intraindividual variations between 0.44 and 0.90 (Asendorpf & van Aken, 1991; Ozer & Gjerde, 1989). An interesting conclusion of these studies is that intraindividual stability is not a result of chance but seems to be associated with positive personality traits such as sociability, emotional control, and conscientiousness, which tend to increase the resilience of individuals to the difficulties of life (Caspi & Shiner, 2006).
In Block’s study , the Q‐sort techniques were used in a pioneering way, and it was established that there were major differences in personality stability rank among individuals. Although the Q‐Sort correlation showed high stability for groups as a whole—0.70 between childhood and late adolescence and 0.50 between this and adult life, these results hid the huge variability between individuals. One interesting conclusion of these studies is that the intraindividual stability is not a random result, but it seems to be associated with positive personality traits, such as sociability, emotional control, and conscientiousness, which tend to increase the resiliency of the individuals facing difficulties in life .
1.5. Personality disorders (PDs) existence and stability before adult life
As we mentioned before, a first question that arises for those who approach the problem of psychopathy in adolescence is to know whether there are personality disorders (PDs) before adult life. In a very influential article, Seagrave and Grisso  draw particular attention to the attention that the physician must have in order not to confuse traits inherent to the development, and therefore transitional, of adolescence itself with the constellation of proto‐psychopath deviant personality traits. That is, to identify true positives, not false positives, as they say. This is a reasonable requirement, and, in our view, it is not being contested. However, as we saw earlier, some authors went even further, stating that there are reasons to believe that in adolescence, there is not even a consolidated personality, and consequently, that there can be no personality disorder. This is the case of Hart, et al. [6, p. 242] when they affirm that there is “no consensus among developmental psychopathologists that personality disorder as a generic class of psychopathology does not even exist in childhood and adolescence” (242). To this extent, it seems to us clear that this matter, whether it is possible to speak of personality disorders before adult life, in what precise terms, from what age, and with what empirical evidence, clinical examination is warranted.
The DSM‐V defines a personality disorder as being
“a persistent pattern of inner experience and behavior that deviates markedly from the expectations of the individual’s culture, is diffuse and inflexible, begins in adolescence or early adulthood, it is stable over time and leads to suffering or injury.” [17, p. 645]
Now, apparently, the DSM recognizes the possibility of personality disorders before adulthood, more specifically in adolescence. Admittedly, there is no clear reference to what stage of adolescence or what possible disorders may occur. But in practice, as Kernberg et al. (2000, p. 6) state that “for personal and theoretical reasons, physicians have been reluctant to diagnose a personality disorder in children and adolescents” [18, p. 6]. These authors state the reasons, more practical than substantial, for this reluctance.
“fear of putting young people prematurely on a negative label that will affect their self‐image and jeopardize them in the future, the refusal by insurance companies to bear health expenses in these cases, on the grounds that such diseases are not officially cataloged, and finally, the conviction, expressed by many professionals, that the personality is not yet sufficiently consolidated to justify such a diagnosis.” [18, p. 6]
Aspects of personal functioning implied in the future constitution of a distinct personality are also early distinguishable, at least from the end of basic schooling, or even earlier, with unique styles of thought and linguistic intelligence, as well as persistence and operability, widely attested by empirical studies, of a style of bonding that, established in childhood, may appear as a striking element for certain so‐called personality pathologies. The question does not, therefore, seem to be whether there are personality disorders before adulthood and even before adolescence. In short, we believe that we have provided evidence that legitimizes us to adopt the point of view of the existence of personality disorders in adolescence, or that these are disorders defined by the lacking developmental framework.
However, these same authors consider the reasons described as unjustified. For them, “PDs in children, as in adults, can be reliably identified, correlated with other disorders of axes I and II, and show a pattern of persistence that makes their impact generalized and strict” [18, p. 14]. And they list a set of other reasons which, in their opinion, support their position: the very early existence (2 and 3 years of age) of self‐consciousness of self, of a feeling of self and of the idea of the other as an empathic individual (resonant to the other).
As several authors argument (i.e., “e.g., see Ref. .”), what needs explanation is not the personality’s mutability, but its opposite, that is, the fact that it presents a considerable degree of stability throughout life.
Sometimes we find in literature references to the extreme volatility of PDs in periods prior to adulthood, especially during adolescence. For example, see Ref. , in a longitudinal study with adolescents from a community population, reports not only a significant incidence of PDs but also a high percentage of cases with spontaneous remission after a short time (approximately 40%). Hart, Watt, and Vincent invoke these results to conclude that there is reason to “doubt the accuracy of the initial diagnosis” [6, p. 242]. Here, we are faced with a problem that escalates the study of PDs in these age groups: the method or procedure used to identify the pathology. Kernberg and his associates  point out that the complexity inherent to the personality construct, a dynamic cluster of traits and components, makes the diagnosis more dependent on the used procedures, at least partially explaining the sometimes enormous differences that we find in studies on the juvenile prevalence of these syndromes. To this extent, it will be much more difficult to diagnose a PD than an isolated trait of the constitutive temperament of that PD, or a pathology of Axis I. To do so, the physician must identify the pattern, often complex and fluid, and often idiosyncratic, of the multiplicity of traits and behaviors that constitute, by definition, the PD, and to interpret the meaning and operational value of these traits and behaviors in the context of the developmental situation in which they occur.
Although there is still much to discover about the formation and stability of the personality and its disorders along the lifespan of individuals, we are not being audacious if we conclude that a basis of evidence has gradually been established which will ensure that the personality itself, and its disorders which may be inherent to it are organized in the majority of individuals before adulthood, not to say that they are organized earlier; or enable to state that the constituent elements of the personality would already be available at a younger age. In this line of thought, for example, in an important longitudinal study that followed 1037 individuals from 3 to 21 years of age, Caspi (2000) showed an impressive continuity of temperamental traits throughout this period. In his opinion, it can start even before 3 years of age, stating that:
“The second year of life may be the crucial dividing line for predicting adult personality differences in adulthood because of the cognitive‐emotional changes that take place during this period. During the second year of life, perceptual and cognitive changes allow the child to acquire the [notion of] permanence of objects and participate in symbolic games. Self‐conscious emotions such as embarrassment and shame also begin to appear at this time These capacities may be necessary for children to form mental representations of their social world and to develop beliefs and expectations that are then confirmed by a reactive and more diverse social environment.” [21, p. 169]
1.6. The nature and stability of psychopathy in childhood and adolescence
The fact that most personality disorders can be built before adulthood does not mean that this is automatically true in the specific case of psychopathy. However, we can discuss with Lynam  that psychopathy is not, in itself, different from other PDs and therefore does not justify a particular a priori reservation regarding this pathology. Or, in the same sense, according to Seagrave and Grisso , the typical traits of psychopathy (impulsivity, a tendency toward deviant behavior, ethical relativism and egocentricity) are precisely those that tend to emerge during adolescence, although with an episodic intensity and transient nature.
In general, the authors agree that there is still a shortage of longitudinal studies on psychopathy.
The authors who affirm the existence of juvenile and adolescent psychopathy do not call into question the need to be especially cautious with the diagnosis of psychopathy applied to children and adolescents. They acknowledge that the label can have serious legal consequences  and that, given the present unresolved controversy over the treatability of this “pathology” [23, 24], these consequences could extend to the clinical domain. It is therefore ethically recommendable that any diagnosis of psychopathy, especially when applied to adolescents, will be supported by scientifically rigorous criteria . Consequently, in general, these authors tend to avoid the designation of psychopaths when referring to pre‐adults, preferring rather that of “individuals with psychopathic traits.” For their part, the arguments invoked to speak of juvenile psychopathy are various. Lynam  coined the term “incipient psychopath” “Fledgling psychopath” to designate children with behavioral problems (CP) who simultaneously exhibited high levels of hyperactive‐impulsive‐attention problems (HIA) and who, in their opinion, were “affected by a virulent variant of behavioral disorder (CD) which will be more appropriately described as fledgling psychopathy” [3, p. 209] and has recently been described by DSMV  as presenting “characteristics necessary for the specifier” “with Limited Prosocial Emotions” with callous and unemotional traits; a search for strong emotions; audacity and insensitivity to punishments. Since then, several researchers have sought to accumulate empirical evidence to determine the characteristics of these young people and whether this category is stable and can be identified with adult psychopathy. It is, in our view, possible to address the multiple reasons invoked to speak of juvenile psychopathy in the following categories:
there is a remarkable stability in temperament and personality traits since adolescence, even since childhood, and in particular, in the callous and unemotional traits and emotional deficit traits that are typical of adult psychopathy;
it is possible to map, according to statistically solid criteria, personality traits of psychopathy (adult and juvenile) in different profiles according to the Big Five model;
there is a laboratory and neuropsychological evidence that suggests that psychopathy is associated with early anatomical and/or physiological dysfunction, possibly with a genetic substrate;
the measurement of juvenile psychopathy by the available instruments shows us that the construct targeted by these instruments has content and factorial validity and is structurally similar to the construct measured by the instruments used with adult psychopaths, especially PCL‐R.
Despite the recent nature of the subject, there are already, as we said, some reflections on these arguments [25–28]). The principal impression after reading these reflections is that, to quote one of them, “the evaluation of psychopathy in children and adolescents is a very important research area, and it is still in its childhood, and our knowledge about nature, stability, and consequences of juvenile psychopathy (…) is very limited “[26, p. 471]. We, therefore, think that it is useful to continue and enhance the research on this issue.
Most of the physicians who are confronted institutionally with heterogeneous groups of juvenile delinquents are liable to recognize, at least in some of these offenders, John Bowlby’s description of fourteen of the 44 boys and girls whom he examined in 1944. The interest of this study will be, above all, in the fact that Bowlby identified in children—under the age of 12—the temperament traits that constitute the main subject of contemporary research on juvenile psychopathy: absence of affective bonds, to react emotionally, precocious and recurrence delinquency, instrumental violence, impulsivity, and superficial charm and deceptive intelligence. And, what is an important aspect, of distinguishing these young people from other aggressive juvenile delinquents, which he calls Hyperthymics . And this is this difference that has, in some way, been one of the key reasons for considering the existence of juvenile psychopathy, isolating it from commonly associated disorders and seeking to give it its own entity.
To conclude, the Klingzell’s study  shows the stability and change of psychopathic traits since childhood. Other authors, through their studies accomplished with community samples, showed that the stability of psychopathic traits could be found since childhood until adulthood, as well as, between the adolescence and adulthood [30, 31].
1.7. Nomological network similar to adult psychopathy
They have been identified by the PCL: YV (The Hare Psychopathy Checklist: Youth Version), and other assessment instruments of adolescent and juvenile psychopathy, some young adults with psychopathic traits, showing a coherent range of characteristics that differentiate them from the remaining juvenile offenders; similar characteristics to those which distinguish adult psychopaths from adults characterized as ASPD non‐psychopaths, for example, callous and unemotional traits [32, 33]; the number of violent acts and the criminal versatility [34, 35]; the recidivism and criminal conduct persistence over the years [36–39]; the preference for the instrumental violence ; and the lack of positive results after the exposure to therapy .
These studies have repeatedly shown that the relationship between the results of psychopathy and the used criteria variables is predicted by the theory, thus reinforcing the idea of the existence of a subgroup of CD with psychopathic characteristics present in the specifier “with Limited Prosocial Emotions” (DSMV). This relationship has been pointed out as strong, persisting even when the PCL: YV or APSD traits constituting the antisocial factor are eliminated from the predictive equations of regression , which shows that the callous and unemotional traits and emotional deficit traits are sufficient to justify, in the current state of our knowledge, the use of the fledgling psychopathy feature as a possible characterizing syndrome of a distinct subgroup of young adults and adolescents with behavioral disorders.
1.8. Psychopathy and personality
In recent years, Donald Lynam has been developing an interesting research program on the relationship between psychopathy and personality structure. The central objective of this program is “not to discuss whether psychopathy is related to personality… [But rather] to gather evidence that psychopathy is personality” [Emphasis of the original]. That is, the basic premise … is that psychopathy can be understood as a particular constellation of basic personality traits … [42, p. 133–134], especially the basic traits recognized by most personality theorists . The reference to this research program is important since, although most researches have been focused on adult psychopathy, Lynam and his colleagues tested the structural model of psychopathy with the structure of personality in samples with adolescents and found that this model was essentially identical to that of adults [22, 27].
Lyman’s and collaborators research program resulted from the convergence of three research strategies:
the performance of a meta‐analysis  on the correlation between psychopathy and personality traits based on studies that allowed this comparison. That is, studies in which in the samples were available psychopathy results obtained with the application of PCL‐R or other specific instruments and results pertaining to basic personality traits resulting from the application of standardized personality measures, namely NEO‐PI‐R from Costa and Mc Crae , Tellegen’s Multidimensional Personality Questionnaire , and Eysenck’s PEN . Although different, these three personality models can be subsumed into a single model, which Lynam calls the four major consensuals (Consensual Big Four), which is composed of the following personality super dimensions: neuroticism, extraversion, pleasure, and conscientiousness.
the translation of the PCL‐R checklist in terms of the language of the structural models of personality, that is, the assignment of each of the 20 items of this checklist to one (or more) dimensions and second‐order traits of those models and determination of its meaning (e.g., item number 1 of PCL‐R, “superficial charm,” was classified as low self‐consciousness, a feature of Neuroticism, so it was defined as N).
the description by recognized psychopathy experts of the typical traits of a psychopath in terms of the language of the structural models of personality  or a non‐theoretical instrument such as Common Language California Child Q‐Sort Version (CLQ) .
These three methods have produced results that converge with each other, allowing a general translation of the personality configuration typical of the psychopath in the language of the structural models of personality . The resulting description varies according to the dimensions: an individual with low Pleasure (A) and high personal antagonism; the psychopath is “egocentric, suspicious, aggressive and does not care about others” [42, p. 139]. The Conscientiousness/Control (C) feature is also associated, in a negative sense, with psychopathy, showing the psychopathic individual as “not able to control himself and to adhere to traditional values and patterns of conduct” [42, p. 139]. Less clear were the results regarding Neuroticism (N) and Extraversion (E), where the relation of psychopathy to these higher‐order personality dimensions seems to be more subtle depending on the second‐order factors to be considered. The psychopath may have high values of hostility and impulsivity (two factors of N), but low values of self‐consciousness (another factor of N), and high values of demand for exciting sensations (an E‐factor) and low positive emotions and cordiality (another factor of E).
The importance of Lyman’s research program in this context in the present chapter is renewed by the fact that, in one of the studies, Common Language Q‐Sort (CLQ) has been used to characterize the “incipient psychopath” . The juvenile psychopath portrait resulting from that study is similar to the one obtained with other methods and for adult population: “The incipient psychopath is extremely low in Agreeableness, extremely low in Conscientiousness and somehow low in Neuroticism” [41, p. 143]. These results were expressed according to the 10 most characteristic verbal descriptions and the 10 less descriptive descriptions of an incipient psychopath, which by its importance are reproduced in Table 2.
|Topic CLQ||Average||DP||Range||FFM Scale|
|11||He tries to blame others for what he does||9.0||0.00||9–9||A− C−|
|22||He tries to make others do what he wants to, manipulating them. He uses his personal charm to get what he wants||8.5||0.76||7–9||A−|
|65||When he wants something, he wants it now. He has difficulty waiting for the things he wants and what he likes||8.5||0.53||8–9||A− C−|
|20||He tries to take advantage of others||8.4||0.71||7–9||A− C−|
|13||He tries to see how far he can go. He goes over the limit and tries to bend the rules as far as possible.||8.3||0.71||7–9||A− C−|
|21||He tries to be the center of attention||8.0||1.07||6–9||A− C−|
|85||He is aggressive||8.0||0.76||7–9||A− C−|
|10||His friendships do not last long, he often changes friends||7.6||1.69||4–9||A− C−|
|91||His emotions do not seem to fit the situation||7.6||0.92||6–9||A− C−|
|93||He is bossy and tries to dominate others||7.6||0.74||6–8||A−|
|76||You can trust him, he is reliable||1.3||0.46||1–2||C+|
|15||He shows concern to what is correct and what is not||1.9||0.99||1–4||A+ C+|
|62||He is obedient and does what he is told to||1.9||0.83||1–3||A+ C+|
|99||He thinks about what he is going to do, he uses his head before he does something or says something||1.9||0.83||1–3||C+|
|9||He establishes solid and intimate relationships with other people||2.0||0.76||1–3||A+|
|67||He plans ahead, he thinks before he does something, he “looks before he jumps”||2.0||0.93||1–3||C+|
|23||He is nervous and fearsome||2.3||0.89||1–3||N+|
|77||He feels insecure, he has a poor opinion of himself||2.4||1.19||1–4||N+|
|2||He is kind and worries about others||2.5||1.60||1–5||A+|
|3||He is a warm person and responds kindly to others||2.5||1.07||2–5||A+|
1.9. Factor structure of juvenile psychopathy measurement instruments
If in some way the structure of juvenile psychopathy proves to be in accordance with the structure of adult psychopathy, it is to be expected that this conformity will also be revealed in the factor structure of the instruments used to measure this “pathology” throughout life. More specifically, measurement instruments for juvenile psychopathy should reveal, for example, when applied to samples with adolescents, a structure semantically similar to that, which is obtained consistently with the classical instruments of adult psychopathy, and especially with PCL‐R. In this respect, if it exists, it will be an additional argument in favor of the existence of adolescent psychopathy, with characteristics that allow us to identify it as an unequivocal precursor of adult psychopathy.
There is currently a considerable range of studies on the psychometric characteristics of juvenile psychopathy measurement instruments. We will examine these instruments in detail later, so here we shall limit ourselves to summarizing their main aspects considered as the most relevant to the present problem. However, the issue is somehow complicated by the fact that the less recent investigations—let us say before 2001—use as a comparative model the traditional bi‐factorial structure of PCL‐R (with two factors, called Factor 1 = Interpersonal/Affective and Factor 2 = Social Deviance) and the more recent tend to be aligned with Cooke et al.  proposal of a three‐factor structure (composed of interpersonal, affective and lifestyle factors) or the proposal of Hare et al. [49, 51], which contemplates a fourth dimension related to the antisocial behavior itself. This fact does not facilitate the comparison between the two types of studies and their integration in the meta‐analysis. It should also be said that the problems of the interpretation of the psychometric properties of the instruments of measurement of juvenile psychopathy are not substantially different from those with their adult counterparts, for example, the problem of the possible hierarchical, or first‐order factorial, nature of the construct, as well as the problem of the true status of the items that operationalize the antisocial dimension (see the controversy between Hare and Cooke and Michie in Recent Handbook of Psychopathy, edited by Patrick [50, 48].
Forth and collaborates  report, in the PCL manual: YV (adapted version of PCL‐R for adolescents), results of validation studies, which show that this instrument presents the same PCL‐R factor structure, in particular, the structure in non‐hierarchical four factors, which is the most recent proposal for this instrument of reference. The confirmatory factorial analysis was based on large samples of 5964 incarcerated adults and in 1631 adolescents. Comparative diagrammatic representations can be analyzed in Hare and Neumann [49, p. 77–78] and show, with small differences, an impressive coincidence in the structural representation between the two groups.
The three‐factor model of Cooke and Michie has also been successfully replicated in PCL: YV [34, 53–55], unlike the original two‐factor model . These results with PCL: YV are especially significant as it is a checklist for adolescents derived from a checklist formulated for adults (PCL‐R).
The similarity between the factor structures of the two instruments is, therefore, an indicator that adolescent psychopathy is a consistent construct.
Factor validation results for other instruments of application in juvenile and adolescent samples have been, however, less conclusive. In the case of APSD , which is also an adaptation of PCL‐R for children between 6 and 12 years of age, some studies report having obtained two factors, similar to those of PCL‐R [2, 58]. In the Frick, Bodin and Barry study , a three‐factor structure was also obtained, but its isomorphism with that of PCL‐R is doubtful. The authors report a factor of callous‐unemotional and interpersonal deficit and another component, called impulsivity and behavior problems (I/CP), which unfolded in two factors, one of which (composed of seven items) reflects interpersonal aspects and which they called narcissism, and the other (consisting of five items) will measure impulsivity. On the other hand, in a study that used a self‐filling variant of APSD, and aimed at adolescents , a trifactorial structure is considered appropriate. Another instrument that has been used with adolescents is the CPS (Childhood Psychopathy Scale), by Lynam , aimed at individuals between the ages of 6 and 17. A revised version, mCPS, and composed of 55 items, was recently presented . This version has an autoresponder variant for teenagers. The structure of this scale is currently still little known, given the paucity of factorial validation studies. Salekin , however, refers to it as a promising instrument for measuring psychopathy. In general, the measurement instruments of juvenile and adolescent psychopathy provide the investigator a mixed background, except for PCL: YV, which seems to measure the same construct as PCL‐R and with the same psychometric structure. The other instruments available differ significantly from PCL and its variants. Although we can say that there is systematically, a factor linked to callous‐unemotional and another to impulsiveness, the status of antisocial behavior items is less clear. One possible explanation for this is that the life history of young people and adolescents is not enough to stabilize this factor, which depends on an accumulation of deviant experiences.
Another possibility is that there is a complex of behavioral traits that, being equally typical and normative of adolescence, confuse the specific antisociality of proto‐psychopaths with the “false positives” spoken by Seagrave and Grisso . In any case, we still have a lot of work to do on the measurement of juvenile psychopathy, without, however, invalidating the allowance for its existence.
2. Final considerations
The dissent among the experts on the main issues is a reality, so it would not be correct of us to accept this task as accomplished. The objections of some relate to matters that are substantive in substance and others with non‐substantive aspects of ethical and legal relevance. In both cases, they should not be ignored. In the case of adult psychopathy, we seem to be moving toward consensus as to its existence as a clinical‐forensic category distinct from the general disorders of antisocial behavior. But even there, many areas of shadow and conceptual indetermination remain. We do not yet know, for example, what is the essential nucleus of psychopathy, whether it has only one or several distinct forms, whether it is preferentially conceptualized as a taxonomic entity or as a dimensional configuration of the personality, and we ignore much about its etiology and about whether or not it has biological origin. These indeterminacies are transposed, for a majority of reasons, into their early or “incipient” manifestations, so it would be unreasonable to imagine that the investigation of psychopathy between pre‐adult populations could simply be discarded from these difficulties. It can also be said that decisive progress in the field of adolescent and juvenile psychopathy will only occur when the study of adult psychopathy advances in the clarification of these fundamental subjects.
But, if this is true, it will be no less true that advances in adult psychopathy studies also depend on the research of its larval or incipient forms. As in biology, the knowledge of adult organisms benefits from what is revealed about the embryonic and growth processes, here again, it will be not only important but also necessary to get to know how this peculiar constellation of personality is established. Therefore, the doubts and reservations about the juvenile and adolescent psychopathy—however legitimate—must yield to the urgency and the need for research. These reserves assume, from the outset, a global character, which will consist of asking whether it is correct to speak, in a general way, of personality disorders in adolescence. We discussed this issue, possibly not as deeply as we would wish, not because it is not so important, but because the deepening would take us too far in a work of this nature, which has obvious limitations of space and ambition. We have quoted the point of view of reference specialists, as Kernberg et al. ., for whom it is legitimate to say, not only that there are personality disorders (PD) in adolescence, but their emergence is probably earlier, dating in many cases of early childhood. However, we need many more studies, especially of longitudinal nature, that will allow us to clarify the forms of manifestation and the degree of stability that PD can present before adulthood.
In the case of psychopathy itself, the situation is perhaps a little better than in the other personality disorders. Psychopathy is, for both intrinsic and historical reasons , a personality disorder par excellence, which has allowed, especially in the last 10 years, a number of significant advances in its incipient manifestations. However, much remains to be clarified, of course, but there is already enough information today that allows us to state, without much risk of error, that we are not investigating a mirage. We have seen that the notion of incipient psychopathy may contribute, like its adult counterpart, to clarify the clinical confusion prevailing in traditional ODD and CD categories [62, 63] We have also seen that there are instruments for measuring juvenile psychopathy—or psychopathic traits, if we prefer a less committed scientific and ethical expression—that present similar psychometric characteristics (such as PCL: YV) or, at least, sufficiently similar to the adult reference instruments (APSD and the mCPS), to the extent that they allow a quantitative rather than just clinical approach to the phenomenology in question. This approach has, for its part, sustained the establishment of a set of theoretical relations between the construct of juvenile psychopathy and other pertinent constructs and variables, which together make up a nomological network similar to that of adult psychopathy. Lynam  has shown that the clinical description by experts of “incipient psychopathy” can be mapped similarly to the adult version in the Big Five terms, and that, at least in phenomenological terms, the two constructs correspond closely. We mentioned studies of genetic heredity and laboratory tradition that show that young people and adolescents with psychopathic traits identified by these instruments exhibit deficits similar to those of their adult counterparts in many of the investigated neurofunctional dimensions.
These considerations are, in our perspective, enough to legitimize the research we have proposed. We cannot assert here adamantly that adolescent psychopathy exists as a stabilized personality disorder and that it assumes an equal (or even sufficiently similar) development to adult psychopathy.