Abstract
Sidney Blatt’s seminal contributions in the domain of personality development, psychopathology, and health rank among the best researched and most empirically supported theories in psychoanalysis. Blatt is known primarily for his two-polarities model of personality development, which he viewed as evolving through a dialectical, synergistic interaction between two fundamental processes across the lifespan: the development of interpersonal relatedness on the one hand, and of self-definition on the other. In this model, psychopathology is viewed as an attempt to find a balance, however distorted, between relatedness and self-definition. Neurobiological research has confirmed the intrinsic dialectical relationship between these two processes in the development of the neural circuits subserving these capacities, a finding with important implications for physical health. Research relevant to these ideas is reviewed, and the influence that Blatt’s approach has had in reintroducing psychodynamic factors into contemporary psychology and psychiatry, as reflected in DSM-5, is discussed.
Faced with the task of summarizing Sidney Blatt’s monumental contributions to our understanding of personality development in relation to mental and physical health, spanning over fifty years of research, one can feel only trepidation. Yet the gist of his views on the relationship between normal and disrupted personality development in relation to mental health can be presented with relative ease and is deceptively simple. Basically, Blatt assumed that personality development evolves through a dialectical, synergistic interaction between two fundamental psychological lines across the lifespan: the development of interpersonal relatedness on the one hand and of self-definition on the other. Various forms of psychopathology are viewed within this approach as attempts to find a balance, however distorted, between relatedness and self-definition, leading to a temporary or chronic emphasis that is excessive, distorted, and often defensive, on one or the other of these two developmental lines.
In what follows I will review the core assumptions of Blatt’s approach to normal and disrupted personality development and the empirical basis of these assumptions. I will also review recent findings demonstrating that the human capacities for interpersonal relatedness and self-definition are closely intertwined at the neurobiological level. I will discuss the influence of disruptions of the dialectical interaction between relatedness and self-definition on physical health, disruptions mediated primarily through impairments in the developing stress system and associated biological systems.
Blatt’s Approach to Normal and Disrupted Personality Development
Although hundreds of empirical studies have led to further theoretical refinements of Blatt’s formulations, demonstrating the progressive nature of the research program that was initiated by these initial formulations (Lakatos 1976), the basic assumptions of Blatt’s two-polarities model of personality development appear to have withstood the test of time. Consistent with basic psychoanalytic assumptions, for Blatt there is no neat distinction between “normal” and “abnormal” or disrupted personality development (Blatt 2008; Blatt and Luyten 2009; Blatt and Shichman 1983). His views also do not assume a neat distinction between normal variations in personality development, “symptom disorders,” and “personality disorders.” As I will discuss in more detail later, mainstream psychology and psychiatry have rediscovered the validity of both of these latter assumptions, and Blatt’s work in this context has had a profound influence, not least on the most recent formulations in DSM-5 (Luyten and Blatt 2016; Skodol and Bender 2009).
Briefly, Blatt viewed personality development as proceeding through continuous dialectical interaction between the capacities for interpersonal relatedness and self-definition—that is, the capacity to establish and maintain, respectively, (1) reciprocal, meaningful, and personally satisfying interpersonal relationships and (2) a coherent, realistic, differentiated, and essentially positive sense of self or an identity (see Figure 1). Influenced by Anna Freud’s concept of developmental lines (1963), Blatt considered these two capacities central to understanding both adaptive and disrupted psychological development and, thus, vulnerability to psychopathology.

Blatt’s two-polarities model of personality development
As reviewed by Beebe and colleagues in this issue, from early infancy onward, experiences of interpersonal relatedness—which include feelings of mutuality and gratifying involvement, but also experiences of incompatibility, misunderstanding, and separation—are thought to contribute to the development of self-definition, which in turn is expected to foster the capacity for interpersonal relatedness, and vice versa (Blatt and Behrends 1987). Hence, human beings are seen as fundamentally interpersonal in nature. In Blatt’s view, even our feelings of selfhood have their roots in experiences of mutuality, as well as incompatibilities and misunderstandings with others, typically beginning in interactions with attachment figures. Influenced by object relations theory and attachment formulations, Blatt viewed experiences of mutuality and incompatibility as instrumental in the need to develop increasingly complex, differentiated, and integrated representations of self and others that influence and guide beliefs and expectations about oneself, others, relationships, and the world (Blatt 2008).
From this perspective, both adaptive and disrupted personality are seen as modes of adaptation that result from variations and disruptions in the synergistic interaction between relatedness and self-definition throughout the lifespan (Luyten and Blatt 2011). These attempts to find a balance between relatedness and autonomy can be more or less adaptive for the individual and/or the environment. According to Blatt (2008), the point of equilibrium that is achieved between relatedness and self-definition depends on a complex set of interactions among psychological, sociological, cultural, historical, and biological factors (see Figure 1). This leads individuals, even within the normal range, to place a somewhat greater emphasis on one or the other of these dimensions. For instance, at least in Western cultures, there would be a tendency for women to place somewhat greater emphasis on relatedness, while men would tend to place somewhat greater emphasis on self-definition. Within collectivistic or interdependent cultures, relatedness would be valued more, while in individualistic or independent cultures, self-definition is considered more normative (Kitayama et al. 1997; Markus and Kitayama 1991; Triandis 2001). Similarly, as reviewed below, parenting styles have been shown to influence the development of a greater emphasis on relatedness versus self-definition.
Blatt and colleagues (e.g., Blatt and Shichman 1983) also used the term anaclitic to refer to individuals whose personality organization is focused predominantly on interpersonal relatedness. This term, taken from the Greek anaklitos, meaning resting or leaning on, was used to refer to Freud’s description (1905) of people with a strong need for relationships resting or leaning on early experiences with caregivers, the Anlehnungstypus der Objektwahl (“anaclitic object choice”). Use of the notion anaclitic was also inspired by Spitz’s description (1945) of a type of “anaclitic depression,” marked by feelings of lethargy and hopelessness, in infants who, separated from their primary caregiver(s), suffered emotional deprivation. In contrast to this notion, Blatt and colleagues (e.g., Blatt and Shichman 1983) used the term introjective to refer to people whose personality organization primarily revolves around self-definition, building on Freud’s notion (1915) that the self is developed through a process of introjection, or identification with significant others.
Anaclitic and introjective personality styles are thus thought to reflect normal variations within personality development, each with its particular experiential mode, preferred forms of cognition, defenses, and adaptation, unique qualities of interpersonal relatedness, and specific forms of self- and object representation (see Figure 2).

Defining features of introjective and anaclitc personality styles
Although these personality configurations represent normal variations, at the extreme they give rise to what consensually is defined as psychopathology. Yet what is important is that even the different forms of psychopathology that are distinguished within a given culture cannot be considered static end-states; rather, they are attempts to maintain a balance, however maladaptive, between relatedness and self-definition. Psychopathology thus reflects dynamic conflict/defense constellations (Luyten and Blatt 2011; McWilliams 2011). According to Blatt, these maladaptive attempts at adaptation typically involve an excessive emphasis on one developmental line and relative neglect of the other (Blatt 2008), leading to a distinction between anaclitic and introjective psychopathology. Anaclitic psychopathology involves, at different developmental levels, a distorted and rigid, one-sided emphasis on interpersonal relatedness; the development of autonomy and self-definition is defensively neglected and/or avoided. Typical examples of anaclitic types of psychopathology include schizophrenia, histrionic, dependent, and borderline personality disorders, and anaclitic depression. In these disorders, relatedness is emphasized at the cost of the development of autonomy and identity; excessive dependency on others is the common theme. Introjective psychopathology, by contrast, is characterized, at different developmental levels, by a distorted and one-sided emphasis on self-definition; an excessive emphasis on the self and autonomy, and a defensive neglect of relatedness, is the common theme in these disorders. Examples are paranoid schizophrenia and paranoid, obsessive-compulsive, self-critical depressive, and narcissistic personality disorders.
Finally, the more excessive the distorted and often defensive neglect of the other line, the more the attempt at finding a balance is likely to fail, particularly in the longer term, and the more it is associated with psychological costs for the individual and his or her environment. As I will discuss in more detail, recent extensions of the model have begun to outline the biological costs associated with disruptions of the normal dialectic between relatedness and self-definition, and their implications for physical health (Luyten and Blatt 2013).
This “double helix theory of personality development” subsequently led to wide-ranging research, particularly as similarities between Blatt’s views and those of other prominent theorists across different theoretical orientations—ranging from philosophy to psychology and psychoanalysis—quickly became apparent (for a review, see Blatt 2008). Subsequent research has shown theoretical and empirical overlap with similar theories proposing that interpersonal relatedness and self-definition can be seen as central organizational features of personality development. In these theories, the two dimensions have been variously referred to as surrender and autonomy (Angyal 1951), communion and agency (Bakan 1966; Pincus 2005), affiliation or intimacy and achievement or power (McAdams 1985; McClelland 1985), relatedness and autonomy/competence (Deci and Ryan 2012; Ryan and Deci 2000), attachment anxiety and attachment avoidance (Mikulincer and Shaver 2007; Sibley 2007), and sociotropy and autonomy (Beck 1983; Clark and Beck 1999). What these theories have in common is that they all propose that adaptive personality development entails a balance between interpersonal relatedness and self-definition, whereas psychopathology involves a temporary or chronic disruption of this balance (Luyten and Blatt 2011). Within attachment theory, for instance, adaptive personality is conceptualized as a balance between relatedness and self-definition, as expressed in low levels of attachment anxiety and avoidance (Mikulincer and Shaver 2007). Insecure attachment styles, by contrast, are thought to represent different types of imbalance between these dimensions. Avoidant attachment, for instance, entails “discomfort with closeness” and “discomfort with depending on others” (Mikulincer and Shaver 2007, p. 87), representing an exaggerated emphasis on self-definition and a defensive avoidance of interpersonal relatedness. Attachment anxiety is typically characterized by “fear of rejection and abandonment” (Mikulincer and Shaver 2007, p. 155), representing an overemphasis on relatedness and a defensive avoidance of feelings of autonomy and self-affirmation (i.e., self-definition).
From the Clinic to the Lab and Back
Blatt’s initial formulations were in part derived from his psychoanalytic experience with patients, but they were also linked to his own life history (Blatt 1974, 2004). Specifically, Blatt recounts how, in the treatment of two patients with depression, he was struck by the very different types of conflict his patients were struggling with. While the depressive experience of one of his patients seemed to revolve primarily around issues of relatedness, such as feelings of abandonment and rejection, the other patient seemed to struggle mainly with issues related to self-definition, such as achievement, guilt, and control. Blatt (2004) goes on to recall that while the depression of the second patient resembled most closely the typical description of depression in much of twentieth-century psychiatry, descriptions of the depressive experience of the first patient were, with some important exceptions, remarkably absent from the psychoanalytic and psychiatric literature. Yet this patient reminded him of seeing his own father cry over his mother’s grave, seemingly feeling desperate and abandoned. This experience, Blatt recounted, probably made him more aware of and attuned to depressive experience related to feelings of abandonment and rejection, and thus is likely to have influenced his theorizing about the importance of such experience.
What perhaps distinguished Blatt most from many of his clinical and theoretical contemporaries was that he decided to put his ideas and formulations to empirical test. Together with colleagues, he developed the Depressive Experiences Questionnaire (DEQ; Blatt, D’Affitti, and Quinlan 1976). This 66-item self-report questionnaire was designed to capture maladaptive expressions of relatedness, as evidenced by high levels of dependency, and maladaptive expressions of self-definition, as captured by the questionnaire’s self-criticism scale. Several other self-report measures assessing these dimensions have emerged since (for reviews, see Luyten, Corveleyn, and Blatt 2005; Zuroff, Mongrain, and Santor 2004); more recently, observer-rated scales of these dimensions have been developed that capture different levels of concern with self-definition and relatedness at different developmental levels. These include the Differentiation-Relatedness Scale (Diamond et al. 2011), the Prototype Matching of Anaclitic and Introjective Personality Orientation (Werbart and Forsström 2014), and the Anaclitic-Introjective Depression Assessment (Rost, Fonagy, and Luyten 2016). Within interpersonal theories, various measures tapping into different facets of relatedness and self-definition have been developed and validated (Locke 2011), while attachment research has focused on interview-based measures, such as the Adult Attachment Interview and the Child Attachment Interview, and experimental paradigms, including the Strange Situation (Ravitz et al. 2010).
A large body of research using these measures has provided evidence supporting the main assumptions of Blatt’s two-polarities model in understanding both normal and disrupted psychological development from childhood to adolescence and adulthood (for reviews, see Luyten and Blatt 2013; Luyten, Corveleyn, and Blatt 2005; Zuroff, Mongrain, and Santor 2004). These views rank among the best-researched, most empirically supported psychoanalytic theories, and have also inspired much research on process-outcome relationships, as Blatt assumed that processes are at work in psychotherapy similar to those operating in normal human development (Blatt et al. 2010).
Overall, this research has shown the particularly destructive nature of self-criticism (Blatt 1995; Shahar 2015), while dependency seems much more a double-edged sword. Associations with psychopathology are typically stronger for self-criticism than for dependency. Studies suggest that this seems due in large part to the fact that highly dependent/anaclitic individuals are able to develop and maintain, though often with much ambivalence, supportive relationships with others, whereas highly introjective/self-critical individuals typically lack the capacity to turn to others, particularly in times of need. These findings again appear to confirm Blatt’s intuition that human beings are fundamentally interpersonal, and that people who show severe impairment in the capacity to turn to others in times of need often seem to fare particularly poorly in life, or at least pay a heavy price for their inability to acknowledge their connectedness to others.
Studies have confirmed that high levels of dependency and self-criticism are both associated with dysfunctional interpersonal transactional cycles (Kiesler 1983), leading to exactly the behavior and reactions by others that the individual fears and tends to avoid, which in turn strengthen the individual’s distorted expectations about him- or herself and others. Highly dependent/anaclitic individuals may be able to generate a positive social environment, but their clinging and demanding relational style often leads to irritation, anger, and resentment in others, which may lead to actual rejection and abandonment, confirming these individuals’ strong fear of being rejected and abandoned. Highly self-critical/introjective individuals, by contrast, tend to be ambivalent, critical, and distrustful of others, because of their fear of criticism and disapproval. Yet this leads them to be perceived by others as cold, distant, and aloof, which confirms these individuals’ belief that others do not like them and disapprove of them.
Studies based on Blatt’s assumptions have amply demonstrated the impact of these dysfunctional transactional cycles on the nature of the therapeutic relationship and therapeutic response (Blatt et al. 2010). Dependent/anaclitic and self-critical/introjective individuals bring very different needs and expectations to therapy, and tend to perceive the therapeutic relationship in personality-congruent ways. As demonstrated by systematic research studies, the identification, articulation, and working through of these transference reactions is crucial in establishing a working alliance and achieving long-term change. Studies in this context, consistent with Blatt’s assumptions, suggest that patients preoccupied primarily with issues of relatedness are responsive mainly to supportive dimensions in therapy. By contrast, patients preoccupied primarily with issues of self-definition are more responsive to interpretive-exploratory dimensions of therapy (Blatt et al. 2010). In this view, successful treatment, regardless of the type of treatment, involves the reactivation of a normal synergistic developmental process in which interpersonal experiences in the therapeutic relationship, as in normal personality development, foster feelings of autonomy and self-definition, which in turn foster interpersonal relatedness.
It is important to point out that problems with relatedness and self-definition thus are transdiagnostic vulnerability factors that are implicated in vulnerability for psychopathology across the lifespan (for reviews, see Blatt 2008; Blatt and Luyten 2010; Egan, Wade, and Shafran 2011; Zuroff, Mongrain, and Santor 2004). These personality factors thus explain, at least in part, the notable heterogeneity within and across different disorders. For instance, in depressed patients, at least four subtypes have been distinguished based on these views, with individuals with clinical depression falling within higher-functioning, less maladaptive, anaclitic Needy Depression and introjective Self-Critical Depression prototypes, and two lower-functioning, more maladaptive, anaclitic Submissive Depression and introjective Dismissive Depression prototypes (Rost, Fonagy, and Luyten 2016).
Similarly, studies suggest that personality disorders can be organized into a cluster focused around issues of relatedness (an “anaclitic” configuration) and a cluster focused around issues of self-definition (an “introjective” configuration) (Blatt and Luyten 2010). The DSM dependent, histrionic, and borderline personality disorders typically fall into the first cluster, while the antisocial, narcissistic, paranoid, schizoid, schizotypal, avoidant, and obsessive-compulsive personality disorders typically fall within the second. These findings have influenced proposals for a radical change in the way personality pathology is organized and assessed in DSM-5, emphasizing the centrality of interpersonal relatedness and self-definition in understanding and classifying personality disorders (Bender, Morey, and Skodol 2011).
Studies based on two-polarities models have also increased our understanding of the intergenerational transmission of vulnerability for psychopathology, suggesting that disproportionate concerns with issues of relatedness and self-definition are transmitted from one generation to the next. This has been demonstrated in studies of children as young as four months (Beebe et al. 2007) and in studies focusing on adolescents (Soenens, Vansteenkiste, and Luyten 2010). A study by Besser and Priel (2005) in a nonclinical sample even showed that excessive preoccupation with relatedness and self-definition was transmitted from mothers to their daughters and granddaughters. Further evidence for the intergenerational transmission of preoccupation with relatedness and self-definition has emerged from attachment research (Verhage et al. 2016).
Two Polarities of Experience, Biology, and Health
Consistent with Blatt’s emphasis on the synergistic interaction between relatedness and self-definition, neurobiological research has suggested considerable overlap in the neural circuitry involved in social cognition with regard to the development of relatedness and the self (Lombardo et al. 2010).
Neural circuits involved in the capacity for relatedness consist primarily of a mesocorticolimbic dopaminergic reward circuit and hypothalamic-midbrain-limbic-paralimbic-cortical circuits (Fonagy, Luyten, and Strathearn 2011; Panksepp 1998; Swain et al. 2007; Vrticka and Vuilleumier 2012). Key neuromodulators in these circuits, including neuropeptides such as oxytocin and vasopressin, seem to be involved in the rewarding features of affiliative behaviors (ranging from parental care to pair-bonding and sexual behavior), in social cognitive capacities underpinning relationships, and in the regulation of behavioral and neuroendocrinological responses to stress (Neumann 2008). Specifically, secure attachment experiences seem to reinforce affiliative behavior, resulting in “broaden and build” cycles (Fredrickson 2001): broadening results from the fact that secure attachment experience encourages explorative behavior; building, in turn, results from the fact that attachment relationships are the “playground” where skills and resources are developed. Hence, in normative development, experiences of relatedness are intrinsically related to the development of feelings of autonomy, competence, and identity (Fonagy and Luyten 2009; Fredrickson 2001; Mikulincer and Shaver 2007). By contrast, disrupted development is associated with a disturbance in the dialectical interaction in the development of neural circuits involved in processing information about self and others.
These assumptions are further reinforced by findings that neural circuits involved in reflecting on self and others are closely intertwined. These circuits include cortical midline structures such as the medial prefrontal cortex, posterior cingulate, precuneus, and temporoparietal junction (Lieberman 2007; Lombardo et al. 2010, 2011; for a meta-analysis, see Northoff et al. 2006).
Consistent with these assumptions, there is now good evidence from research in both humans and animals that attachment experience plays a key role in programming the development of the hypothalamic-pituitary-adrenal axis, the main human stress system underpinning our capacity to deal with conflict and adversity, and in associated biological systems such as the immune system and pain-processing systems (Lupien et al. 2009; Luyten et al. 2013; Sbarra and Hazan 2008). These early programming effects are currently thought to explain, in part, vulnerability to psychiatric disorders as well as to functional somatic disorders across the lifespan (Lupien et al. 2009). Consistent with these views, research has shown how both highly self-critical/introjective and highly dependent/anaclitic individuals are more prone to develop a host of functional somatic problems and disorders, with most of the evidence again being for the role of self-critical/introjective features (Luyten et al. 2013; Shahar 2015). Disruptions in stress regulation and associated biological systems have been shown to mediate these relationships.
Conclusions
In contrast to more static symptom- or disorder-centered approaches, Blatt’s two-polarities model fundamentally proposes that psychopathology reflects attempts to achieve a measure of stability in response to developmental disruptions by becoming preoccupied, at different developmental levels, with one or the other of the developmental dimensions of interpersonal relatedness and self-definition (Blatt 2008; Luyten et al. 2012).
Although Blatt’s views have undergone a series of reformulations over the years, the basic formulations of his two-polarities model continue to provide a productive theoretical, empirical, and clinical framework for conceptualizing the connections between normal and disrupted personality development, vulnerability for psychopathology, mental and physical health, and responsiveness to psychotherapy. Above all, these ideas demonstrate the ability of psychoanalytic thinking to influence mainstream psychology and psychiatry, evincing a dedication to systematic empirical research while remaining solidly rooted in psychoanalytic theory and experience.
Footnotes
Faculty of Psychology and Educational Sciences, University of Leuven, Belgium; Research Department of Clinical, Educational, and Health Psychology, University College London.
