Nature and Assessment of Personality Pathology and Diagnosis

The DSM-5 (1), published nearly 10 years ago, introduced the Alternative DSM-5 Model for Personality Disorders (AMPD) in section III (2–4). In contrast to the personality disorder classification system in section II of the DSM-5 that has retained the traditional 10 polythetic categorical (discrete) personality disorder diagnoses, the AMPD requires clinicians to make three determinations. First, the clinician assesses criterion A, the level of personality functioning (LPF), which is used to determine a patient’s severity of impairment in personality functioning, defined as disturbed self-functioning (identity and self-direction) and interpersonal functioning (empathy and intimacy). Second, the clinician evaluates severity in the five criterion B maladaptive trait domains (negative affectivity, detachment, antagonism, disinhibition, and psychoticism). In the third and final step, which is optional, the clinician considers whether criterion A and B characteristics correspond to any of the six personality disorders that were retained in the AMPD. This step is taken only when the clinician is interested in diagnosing a traditional, categorically defined personality disorder.

Similar nosological advances in terms of the dimensionalization of personality disorder diagnosis are reflected in the new ICD-11 diagnostic system (5, 6). The ICD-11 removed all categorically defined personality disorders (except for a borderline pattern qualifier, which was retained to support service transitions organized around the construct of borderline personality disorder). Similar to the AMPD, the ICD-11 includes a general severity criterion, defined as impairments in functioning of aspects of the self or problems in interpersonal functioning that manifest in maladaptive (e.g., inflexible or poorly regulated) patterns of cognition, emotional experience, emotional expression, and behavior. After determining the level of severity of impairment in self- and interpersonal functioning (mild, moderate, severe), the clinician has the option (but is not required) to consider severity levels on five trait domains, referred to as trait domain qualifiers.

The fundamental advancement of the AMPD is that it articulates a core dimension of personality pathology that is common to all manifestations of personality pathology and is represented in the LPF. Criterion B traits provide the so-called flavor of personality disorder dysfunction. Similarly, ICD-11 defines a general severity criterion as necessary (and sufficient) for the diagnosis of personality disorder; evaluation of trait domains is optional. This reformulation of personality disorder, with its shift away from types to define the common core of personality function, has been welcomed by clinicians and researchers for several reasons (7). First, the dimensionalization of personality pathology is consistent with empirical evidence that places personality function on a continuum (dimension) rather than in a category (type, taxon) (8). Second, and consistent with the overall goal of the current article, the reintroduction of the idea of self- and interpersonal dysfunction as the core and common feature of personality disorder puts back into focus the meaning of personality as defined more than 100 years ago—that is, the subjective experience of what it means to be human (8)—and confirms long-standing psychodynamic and psychoanalytic conceptualizations of personality psychology as articulated in Otto Kernberg’s psychoanalytic model of personality structure (9). Indeed, Kernberg’s work played a major role in the development of the Level of Personality Functioning Scale, of which two of the main architects were psychoanalysts (10).

In this article, we pay tribute to the theoretical work of Kernberg in the development and evolution of the object relations theory of personality pathology by demonstrating his contribution to the current understanding of the nature and assessment of personality pathology and diagnosis. We discuss differing views that have emerged regarding the meaning of LPF, in order to highlight the importance of Kernberg’s theory. Next, we discuss Kernberg’s theory and focus on its significance to recent nosological advances. We highlight two important features of Kernberg’s theory: first, a definition of personality that goes beyond mere signs and symptoms to include structural motivational components, in the domains of self- and interpersonal functioning, that are common to all personality manifestations and that fulfill an intrapsychic, organizing function; second, identity formation and consolidation as the central and ultimate essence of healthy personality function. We show that these cornerstone features of Kernberg’s theory, first articulated more than 50 years ago, represent an idea whose time has finally come.

Differing Views Regarding the Meaning of LPF, the Entry Criterion for Personality Pathology

The AMPD in many ways represents a consensus system for the diagnosis of personality disorder. Proponents of a categorical approach to the diagnosis of personality pathology remain able to determine, during the third step in the AMPD, the presence or absence of specific disorders, for instance, borderline personality disorder. Trait researchers, who are less concerned with categorical diagnosis, are satisfied by the trait representation in criterion B of the AMPD. Proponents of a trait approach are further gratified by the fact that criterion A resolves a fundamental problem in using traits to diagnose personality pathology—that is, the fact that extreme traits themselves are not necessarily pathological and may therefore not be sufficient to define disordered personality (11, 12). Although a clinical cutoff point can be derived, doing so would catapult personality disorder diagnosis back into an undesired categorical (yes or no) framework. For some trait psychologists, criterion A therefore allows the assessment of whether a person is struggling with work and love (see Freud, as quoted by Erikson [13]). If dysfunction is detected, the trait psychologist concludes that extreme traits have interfered with functioning, indicating that a diagnosis of personality disorder is warranted. Traits from criterion B then provide the roadmap for intervention (14).

For other trait psychologists, an evaluation of dysfunction with criterion A is entirely redundant (15, 16). These authors argue that dysfunction indicated by criterion A is fully accounted for by the traits from criterion B (e.g., deficits in empathy from criterion A overlap with the maladaptive trait of callousness from criterion B, and problems in relationships are reflected in both interpersonal dysfunction [criterion A] and detachment domains [criterion B]). Moreover, according to this view, because criterion A identifies nothing more than dysfunction (or level of severity), some researchers have argued that criterion A is redundant. In this view, dysfunction is adequately represented in a higher-order dimension of severity—the so-called psychopathology factor (p factor) (17), which has been shown to account for the high covariance among psychopathology dimensions of internalizing, externalizing, and psychoticism spectra and is represented in the “hierarchical taxonomy of psychopathology” model as the highest level of a hierarchical covariance structure of psychopathology (18). The five major spectrum dimensions (internalizing, thought disorder, disinhibited externalizing, antagonistic externalizing, and detachment) from which the p factor emerges closely correspond to the trait domains of AMPD’s criterion B (negative affectivity, psychoticism, disinhibition, antagonism, and detachment). Because the p factor, which indexes overall severity across all dimensions of psychopathology, can be captured thus, a separate severity criterion (i.e., criterion A) becomes unnecessary.

There are obvious problems with considering criterion A as a mere indicator of dysfunction in work and love or with discarding it—see Sharp and Wall (7) for a full discussion of arguments against these propositions. First, the use of criterion A to indicate general dysfunction is problematic, because it suggests that criterion A has nothing to do with personality per se and merely indexes general disability. If this is so, then why retain a criterion as part of personality disorder classification that has nothing to do with personality in particular? Interference with general functioning is a hallmark of all psychopathology and is not specific to personality function. Second, discarding criterion A in favor of the p factor to capture general dysfunction raises the question of how high a p factor score needs to be to indicate dysfunction. Presumably, a cutoff score can be derived, but how is this measure different from measures of general disability or dysfunction as captured in, for instance, the traditional axis V Global Assessment of Functioning (GAF) score? The validity of the GAF has been called into question, resulting in the removal of the measure from the DSM system. Moreover, research has shown a distinction between measures of criterion A and indices of general function in both adults (19) and adolescents (20). Third, how should the p factor be assessed in an individual patient, given that its derivation in a covariance structure is a research method that is based on factor analyses among large samples of individuals? Use of total scores on a psychopathology measure or disability measures is an option, but how does such a score relate to personality dysfunction in particular? This question brings us to a fourth question that directly points to the importance of Kernberg’s theory to a full understanding of the nature and assessment of personality pathology: does variation in maladaptive trait function, which maps onto the internalizing, externalizing, and psychoticism spectra (18), adequately capture personality functioning? In other words, does information about trait function tell us who a person is, or is additional information beyond signs, symptoms, and manifestations needed to fully describe a person?

A fifth and final challenge to a purely trait conceptualization of personality pathology emerges when considering the development of personality pathology. As we have discussed in detail elsewhere (7, 21–24), a child’s position on any dispositional trait dimension (criterion B) can be readily identified already in infancy. Although research suggests a normative increase in maladaptive personality traits in adolescence, followed by a normative decline thereafter, a child’s position on criterion B dimensions relative to his or her same-age peers remains stable throughout development. Furthermore, although even infants may have extreme temperament measure scores that are indicative of maladaptive trait function, extreme traits themselves do not necessarily denote personality disorder. What, then, would compel a clinician to diagnose a personality disorder for a 14-year-old? Would high levels of impulsivity, emotional reactivity, and disagreeableness suffice? Or is the clinician looking for another developmental milestone that should be achieved but is not manifesting in the expected way?

In this article, we contend that criterion B is not enough to tell us who a person is and, by implication, the extent to which an individual manifests healthy personality functioning. We suggest that to adequately assess personality function, we need to understand something about an individual’s subjective experience of her or his self in relation to others, as captured in the personality disorder entry criterion of both the AMPD and ICD-11. Personality emerges from the subjective experience of the self and from intersubjective experience between oneself and others (25). We also contend that criterion B is not enough to explain the onset of personality disorder, and that an additional developmental mechanism—identity consolidation—is needed to determine whether or not extreme traits indicate personality dysfunction. As we will show, the seeds for these assertions were planted more than 50 years ago by Kernberg’s theory of the structure of personality and by the central role played by identity function within this structure.

Kernberg’s Object Relations Theory: An Idea Whose Time Has Finally Come

LPF as the Structural Motivational Components That Fulfill an Intrapsychic, Organizing Function

Kernberg’s object relations model of personality organization (26) postulates that affectively charged early experiences in caregiver-infant interactions are gradually internalized as object relations dyads of self-other representations and provide the fundamental building blocks of personality organization. In early development, these object relations dyads are unintegrated: representations of self and other are not well differentiated, and there is a split between all-good and all-bad (persecutory) experiences. In typical development, object relations dyads become more integrated, resulting in a healthy personality structure, of which the hallmark feature is a coherent, integrated, and stable identity and a capacity for reciprocal and healthy relationships with attachment figures—in AMPD terms, adaptive self- and interpersonal functioning. In contrast, in pathological psychic functioning, integration and differentiation of self-other representations are delayed, partially achieved, or absent, resulting in identity diffusion and maladaptive interpersonal relationships (9, 27, 28)—in AMPD terms, maladaptive self- and interpersonal functioning.

Kernberg’s theory of personality function is a structural one, in the tradition of psychoanalytic theory. With its focus on internal representations of self in relation to others, personality is given a meaning-making function that organizes subjective experience and interpersonal functioning. This means that personality does something; it has an active intrapsychic function that organizes subjective experience and behavior in relatively predictable ways. In contrast, traits happen to people; they are dispositions that are biologically grounded in temperament. Through interaction with the environment, traits become the consequences or manifestations of intrapsychic function. In this sense, Kernberg’s object relations theory both describes and explains personality function and can be understood as a process theory of personality functioning (24). Internal object relations (representations of self and others) can be thought of as schemas that are activated in particular situations and then enacted. In this sense, representations are not mere historically valid representations of early attachment relations but are constructions of one’s relational reality in the present moment. Akin to mentalization-based theory (29), Kernberg’s theory suggests that the decoupling between what is in the mind of the individual and what is present in the real world is the hallmark of healthy personality functioning.

A structural diagnosis in the Kernbergian sense involves an assessment of the severity of personality pathology (30). Four levels of severity in personality organization are described: normal, neurotic, borderline, and psychotic (30). Borderline personality organization is further characterized on a dimension, ranging from low in organization (associated with more pathology) to high in organization (associated with less pathology). An assessment of identity consolidation (or, conversely, diffusion), such as the Structured Interview of Personality Organization–Revised (31), can be used to distinguish between organization severity levels (26, 30). Like the AMPD criteria, each level represents a diagnostic prototype (or mode of psychic functioning) that conveys information about severity and prognosis of pathology, guides treatment planning, and facilitates communication among clinicians (32). Table 1 provides the adaptive and maladaptive end points of the levels of severity across aspects of personality functioning according to Kernberg’s model (33).

In this model, identity consolidation represents the sum of healthy object relations, mature defenses, integration of aggressive impulses, and owned and understood moral values. For Kernberg, the term “severity” does not mean dysfunction or disability (e.g., GAF scores or total p factor score) as it does for trait psychologists. Instead, in Kernberg’s theory, severity is defined as an individual’s subjective experience of themselves and their attachment (object) relationships, the nature of their defensive operations, and the stability of their reality testing. Therefore, severity is not the consequence of personality disorder but its cause or its source (7)—that is, the structural motivational components in the domains of self- and interpersonal functioning that fulfill an intrapsychic, organizing function. It follows that a more effective strategy for treating personality pathology would be to target underlying processes that organize all personality disorders—that is, maladaptive self-other representations (object relations), or, in AMPD terms, maladaptive self- and interpersonal functioning (criterion A). In contrast to trait-based solutions to treating personality pathology (14), which promote the targeting of individual symptoms or behaviors (e.g., emotion dysregulation, self-harm), Kernberg suggests treating its cause, because in his model, treating the manifestations of personality pathology rather than its cause is associated with “risk of treatments devolving into repetitive cycles of chasing symptoms or unfocused pursuit of psychological exploration” (33).

The shift toward targeting the common core (criterion A) of personality pathology that encompasses all manifestations or personality disorder flavors (criterion B) is supported by factor analytic research that has demonstrated that borderline personality disorder, as defined in section II of the DSM-5, appears to load exclusively onto a general factor of personality pathology, whereas other section II personality disorders appear to represent specific factors or maladaptive trait constellations (34–36). Thus, section II–defined borderline personality disorder (like criterion A) may represent the common core features shared by all personality pathologies (22, 23, 37, 38). This conclusion makes sense when considering the fact that, compared with other personality disorders, which were often reduced to purely behavioral manifestations of personality pathology in the DSM-IV, the diagnosis of borderline personality disorder retains explicit criteria reflective of intrapsychic, maladaptive self- and interpersonal functioning (21). The extent to which section II borderline personality disorder fully captures the general factor of personality pathology and, therefore, criterion A remains an empirical question. However, given their suggested overlap (at least in adults), it is possible to argue that borderline personality disorder, the general factor of personality pathology, or criterion A represents an index of increased severity in psychopathology (12, 22, 23, 39), somewhere along the pathway between the internalizing and externalizing spectra and psychoticism—exactly as Kernberg predicted (9, 17).

Identity Formation and Consolidation as the Essence of Healthy Personality Functioning

Because Kernberg gives identity function a coalescing role as the sum of healthy object relations, mature defenses, integration of aggressive impulses, and healthy moral functioning, his understanding of the nature of personality pathology also provides an answer to the question of why personality pathology has its onset in adolescence. As Erikson (13) pointed out more than 50 years ago, one of the major tasks of adolescence is to establish a coherent and integrated sense of self, which facilitates adult role function. To achieve this, an adolescent must successfully navigate the process of becoming a separate individual while remaining connected to others—most notably to parental attachment figures. During adolescence, significant developmental advances in the meta-cognitive capacity for self-reflection and for mentalizing facilitate the ability to, for the first time, ask questions such as, “Who am I?” “How do I want others to view me?” and “How do I fit into the larger social world?” (22, 40, 41). Although some of the developmental building blocks of these capacities are observable during preadolescence, it is not until adolescence that unintegrated aspects of self-functioning (e.g., self-concept, self-esteem, self-directedness, self-reflection) begin to coalesce (or bind) into a unidimensional continuum that ranges from healthy to unhealthy personality functioning (21, 42). At this stage, for the first time, an adolescent can symbolize, perceive, and organize experience into some meaningful relation to the self. In the case of personality disorder, these processes do not bind into a coherent sense of self, and personality pathology ensues. It is critical to understand that without consideration of the structural motivational components, in the domains of self- and interpersonal functioning, that fulfill an intrapsychic, organizing function, trait function alone cannot distinguish whether an adolescent is on the way to developing personality pathology or simply struggling with normative developmental issues or internalizing or externalizing problems.

Conclusions

In this article, we have highlighted two important features of Kernberg’s object relations theory that have implications for assessment and diagnosis of personality pathology, specifically in light of current confusion in the field regarding the entry criterion for personality disorder in the current diagnostic approaches. We have shown how Kernberg’s definition of personality promotes a view of personality function that goes beyond description of mere signs and symptoms (criterion B or section II personality disorder diagnosis) to include structural motivational components, in the domains of self- and interpersonal functioning, that are common to all personality manifestations (criterion A). These components fulfill an intrapsychic, organizing function that matures during adolescence and is represented in the coalescing or binding of identity formation and consolidation. Any psychotherapy should therefore include a focus on scaffolding identity function, as, for example, transference-focused psychotherapy ultimately aims to do.