Personality AnalysisPrior to beginning work on this assignment, review Chapter 3 in your textbook and the HumanMetrics Jung Typology Test website, and read the Choca (1999), Paris (2005), and Westen (

K W W S G [ G R L R U J 0 D N L Q J ' L D J Q R V L V 0 H D Q L Q J I X O ( Q K D Q F L Q J ( Y D O X D W L R Q D Q G 7 U H D W P H Q W R I 3 V \ F K R O R J L F D O ' L V R U G H U V H G L W H G E \ - : % D U U R Q & R S \ U L J K W k $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q $ O O U L J K W V U H V H U Y H G & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q CASE FORMULATION AND PERSONALITY DIAGNOSIS: TWO PROCESSES OR ONE? DREW WESTEN Clinical formulation and diagnostic formulation often seem like en- tirely separate, if not contradictory, enterprises, particularly with respect to the assessment of personality and its pathology. Clinical case formulation always presupposes a theory of personality because the questions one asks and the hypotheses one forms about a patient’s personality depend on what one thinks personality is and how it relates to overt symptomatology. The categories and criteria embodied in the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV; American Psychiatric As- sociation, 1994), by contrast, were deliberately selected to be as theory neutral as possible. As a result, clinicians, who always operate from theory, often find them irrelevant to clinical formulation and practice. Does it matter whether a patient meets criteria for histrionic personality disorder or only manifests four of the requisite criteria?

In this chapter I argue that one of the problems with the current diagnostic system is precisely that it separates the processes of formulation and diagnosis, that this drives a wedge between clinical practice and re- search, that this need not be the case, and that quantified clinical judg- 111 & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q ments can in fact be translated into clinically and empirically valid per- sonality diagnoses. I begin by addressing some of the costs and benefits of diagnosis, focusing on the diagnosis of personality and personality disorders. I then suggest that the link between case formulation and diagnosis is functional dulgnosis (i.e., assessment of discrete but interacting personality functions) and suggest three questions that organize a systematic case for- mulation: What are the person’s motives; psychological resources; and ex- perience of the self, others, and relationships? A fourth related question is, How is each of these aspects of the person’s personality developed? I con- clude with a brief discussion of a psychometric procedure for translating reliable clinical judgments about personality functioning into diagnoses, which can be adapted for clinical practice in a way that systematizes and organizes clinical thinking rather than overrides it for the purpose of ar- riving at what may seem like an arbitrary diagnostic categorization. WHY DIAGNOSE? THE COSTS AND BENEFITS OF DIAGNOSING PERSONALITY I begin with a case example, which I use throughout the chapter to illustrate some of the relevant issues. Throughout the chapter, the case description and formulation are indented, so that by the end of the chapter the reader can assess the adequacy of the model of case formulation being proposed by rereading the material in italics as if it were an initial evalu- ation summary. For simplicity, I describe the patient at the beginning of treatment. The patient, Mr. D., was a man in his early 20s who came to treatment for lifelong problems with depression, anxiety, and feelings of inade- quacy. Mr. D. was a kind, introspective, sensitive man who nevertheless had tremendous difficulty making friends and interacting comfortably with people. He was constantly worried that he would misspeak, would ruminate after conversations about what he had said and the way he was perceived, and had only one or two friends with whom he could interact comfortably. Sex for Mr. D. was fraught with conflict. He was in a 2-year relationship with a woman who was emotionally and phys- ically distant, whom he saw twice a month and with whom he rarely had sex. Before her, his sexual experiences all had been anxiety pro- voking and short-lived in every sense. His associations to memories of these sexual experiences were replete with classically Freudian imagery, such as his fantasy that he would “accidentally” touch the woman’s anus and be repulsed, that her vagina seemed “torn up,” and that his desire was really for a man. Mr. D. came from a working-class family in Boston and had lost his father, a police officer, when his father was killed in a gunfight when Mr. D. was 4 years old. He was reared by his mother and later a stepfather, who seemed relatively kind and benign. 112 DREW WESTEN & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q He described his mother as basically loving, but she herself had a his- tory of depression that seemed more chronic than episodic.

Using the DSM-IV, Mr. D. would not receive a personality disorder (PD) diagnosis, even though he clearly had enduring, maladaptive ways of thinking, feeling, and behaving (i.e., the kinds of personality problems that require therapeutic attention). On Axis I he meets criteria for dysthymic disorder, has subclinical symptoms of an anxiety disorder, and could be diagnosed with a male erectile disorder. On Axis I1 he would come close to meeting criteria for depressive personality disorder, which is not an of- ficial diagnosis. He has some avoidant features but clearly does not meet criteria for that disorder. Using the two axes, then, one can capture much of his depression, some of his anxiety, some of his social anxiety, and a description of some of his sexual problems. The interconnections among these problems, however, cannot be captured; they appear as discrete symp- toms. His chart diagnosis would be nondescriptive:

Axis I: dysthymic disorder; male erectile disorder; r/o anxiety dis- order not otherwise specified; Axis 11: deferred; Axis Ill: deferred; Axis IV: social isolation, job dissatisfaction; Axis V (Global Assessment of Functioning): 65. I believe this is a relatively typical case in terms of what the DSM-IV captures and fails to capture. The Problems of Diagnosis Diagnosis, particularly as embodied in the DSM approach, has a num- ber of problems, costs, and limitations. One with which every introductory psychology student is familiar is the problem of labeling. The classic version of this concern was the labeling theories of the 1960s, which focused on the stigma of psychiatric diagnosis, notably schizophrenia. A more contem- porary concern is that labels activate schemas, and schemas carry conno- tations that can lead clinicians to see what they expect or to react affec- tively to patients diagnosed with disorders for which they have particular countertransference problems (e.g., the reactions of many clinicians to pa- tients with borderline personality disorder [BPD]). Additionally, diagnosis today is frequently used in the service of anxiety reduction by psychiatric residents and other mental health professionals who are not being ade- quately trained to understand their patients’ dynamics and instead are com- pensating by focusing their attention on finding the right category into which to place their patients. These problems associated with psychiatric labeling, however, point more to problems with clinicians and training than with diagnosis per se. They suggest that as clinicians we need to be CASE FORMULATION AND DIAGNOSIS 113 & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q aware of schematic biases and our issues with particular kinds of patients, address difficulties we might have in sitting with “not knowing,” and avoid training programs that teach about categories instead of people.

Other problems, however, are more serious. One is the “so what” question. Suppose we decide that Mr. D. has a dysthymic disorder, and perhaps make the diagnosis of depressive personality disorder even though it is currently only in the appendix to the diagnostic manual. Clinically, what does this add to the knowledge, garnered in the first 5 min of the first interview, that he has been depressed most of his life? Does it guide us psychotherapeutically? Does it suggest a particular medication? Suppose, instead, we decide that he has an avoidant personality disorder with de- pressive features rather than a depressive personality disorder with avoidant features. So what? These are the kinds of diagnostic questions on which most PD research is currently focused, which is probably why most clini- cians, sadly, do not read PD research.

Another related problem is the issue of comorbidity. The DSM ap- proach, following a model that does pass the “so what” test in other areas of medicine (e.g., knowing that a patient has hypothyroidism rather than anemia has clear treatment implications), places individuals into catego- ries. The problem for the classification of PDs is the high comorbidity of PDs both within and across the two axes. With respect to Axis I comor- bidity, empirically, a substantial percentage of patients with Axis I1 disor- ders are comorbid for Axis I pathology and vice versa (Green & Curtis, 1988; Shea, Glass, Pilkonis, Watkins, & Docherty, 1987). The whole notion of Axis I-Axis I1 “comorbidity” may be largely artifactual, brought about by an artificial distinction between one axis originally intended to be more episodic and biogenic and the other more chronic and psycho- genic. Unfortunately, nature was not so kind as to separate episodic, bio- logical disorders from chronic, psychological or “functional” pathology.

Schizophrenia and bipolar disorder are the best examples of disorders with high heritability, but schizophrenia is often chronic, as are some forms of bipolar disorder, and relapse in each is related in part to psychosocial var- iables such as the amount of hostile criticism and overinvolvement patients receive from their families (Hooley, 1987; Miklowitz et al., 1991). Fur- thermore, many of the PDs undoubtedly have biological diatheses, partic- ularly schizotypal PD, which appears to be best categorized as a schizo- phrenic spectrum disorder and is found in the biological relatives of schizophrenic patients.

In this respect, consider Mr.

D. Does it really make sense to scatter his symptomatology over two axes, like ashes over the Grand Canyon? All the problems with which he presented are chronic, but the extent to which they reflect biological or environmental etiological factors is unclear. Mak- ing matters more complicated, many aspects of his pathology are not rep- resented on either Axis, such as his sexual conflicts, his passivity, and his 114 DREW WESTEN & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q chronic feelings of inferiority that do not appear to be part of any particular PD and are found in many patients who have personality problems not severe enough to merit an Axis I1 diagnosis. Indeed, this latter point rep- resents another clear problem with Axis 11, which fails to include or cat- egorize a large percentage of psychotherapy patients with “neurotic” per- sonality pathology that slips between the axes (for empirical data, see Westen, 1997a). In a recent study, Westen and Arkowitz (1998) found that fully 60% of patients treated in clinical practice for enduring maladaptive personality patterns cannot be diagnosed on Axis 11.

Within Axis 11, comorbidity is even more of a problem. Research consistently shows that most patients with PDs fit criteria for multiple disorders (Bell & Jackson, 1992; Morey, 1988; Oldham et al., 1992), so that if a patient has any PD, he or she likely receives three or four PD diagnoses, at least with the research instruments widely in use. This has led some observers (Widiger & Frances, 1994) to suggest a shift from cat- egorical to dimensional diagnosis. This could be accomplished by giving the patient a rating for each PD, leading to a diagnostic profile rather than a categorical diagnosis (e.g., on a scale from 1 to 7, Mr. D. would receive a 6 on depressive PD, a 3 on avoidant PD, a 1 on BPD, etc.). Alternatively, as Widiger and Frances and others have proposed, Axis I1 could be aban- doned in favor of rating patients using the Five Factor Model (FFM) of personality (Goldberg, 1993; John, 1990; McCrae & Costa, 1990). The FFM was derived from self-reports and distinguishes five dimensions of per- sonality on which individuals vary: neuroticism, extraversion, conscien- tiousness, agreeableness, and openness to experience. From this perspective, people with PDs are nothing but individuals who fall on the extreme ends of these dimensions. Mr. D. would no longer receive a diagnosis of de- pressive PD with avoidant features; instead, he would score high on neu- roticism and introversion.

From a clinical perspective, neither the current categorical model nor the dimensional alternative seems to provide an adequate solution to the problem of personality diagnosis. In fact, each seems to confound a sub- stantive issue with a scaling issue. clinically, the question is not, Does the patient cross the threshold for avoidant PD? or How high does the patient score on neuroticism? The relevant clinical question is, Under what circumstances do which cognitive, affective, motivational, and be- havioral patterns and their interactions get triggered in ways that lead to distress for the patient or those around him or her? As I show later, this is a functional question, not a scaling question, and it is one that neither Axis I1 nor the FFM (particularly with its reliance on self-reports) is pre- pared to answer. It is a question that deals with the dynamic interaction of psychological processes, intrapsychically and intersystemically. Both cat- egorical and dimensional diagnostic systems share a major problem of trait theories criticized years ago by Mischel (1968) with which personality psy- CASE FORMULATION AND DIAGNOSIS J 15 & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q chologists have wrestled for nearly three decades (e.g., Funder & Colvin, 1991; Kenrick & Funder, 1988), namely that they fail to specify the elic- iting conditions for personality processes and instead describe traits as qual- ities that are relatively fixed over time and across situations. Clinically, many patients have problems with authority, but these do not emerge with cats, dogs, peers, or subordinates and cannot easily be summarized with adjectives on a checklist or with categorical diagnoses.

The Benefits of Diagnosis These are some strong arguments against diagnosis, particularly as instantiated in the DSM-lV. Diagnosis also, however, has considerable benefits. One is communication between clinicians. To describe a patient with a 10-page discursive evaluation summary is not parsimonious, nor can clinicians learn about more general processes typical of certain types of patients (e.g., those with BPD, antisocial PD, anorexia, etc.) without some form of categorization. Elimination of diagnosis would result in articles with titles such as, “Developmental History in Patients Who Show Evidence of Splitting, Difficulty Maintaining Relationships, Self-Mutilation, Inconstant Representations, Poorly Modulated Affect, and Sundry Other Symptoms.” Clearly, diagnosis allows communication among mental health profession- als about patients who share certain features, however fuzzy the categories are around the edges. A second benefit has to do with communication with patients. As clinicians, we rarely present our patients with a diagnosis. Rather, if we provide initial diagnostic feedback at all (which I believe we should), we offer a brief narrative, clinical formulation at the end of an evaluation period, which describes what we have seen and what we think might be helpful. For example, after spending three sessions with a patient with BPD, I might say something along the following lines: It seems like we’ve identified some central issues for us to work on.

One is that you often feel overwhelmed, like your feelings are out of control. When you feel this way, sometimes you’ll break off a relation- ship that you’ll later wish you hadn’t, sometimes you’ll cut yourself, and sometimes you’ll drink too much. A second issue is that deep down, you seem like you don’t really trust that people will love you and treat you well. And this seems related to your feelings about yourself-when you feel bad about yourself, you feel totally bad, like there’s nothing about you but badness and that you don’t deserve any better. This can be useful to patients because it takes what feels like a morass of feelings, symptoms, and concerns and puts them into a manageable, co- herent form and lets them know that the clinician has some understanding of them and how to proceed. 116 DREW WESTEN & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q On some occasions, however, actually telling a patient his or her diagnosis can be therapeutically indicated. This is often the case with pa- tients with posttraumatic stress disorder, who frequently fear they are going crazy and have no idea that their symptoms are a typical, understandable, and treatable response to psychological trauma. Although using a diagnos- tic label is rarely indicated with PD patients, who will generally feel labeled rather than understood (and for good reason), in some cases giving the patient a diagnosis can be helpful. When I was nearing the end of my internship many years ago, I was assigned a severely disturbed borderline patient for evaluation and brief treatment (because I was leaving the agency). She presented with a foot-high stack of psychiatric records since the time of her initial contacts with mental health professionals as a run- away 20 years earlier. I knew she had written away for all prior records and that she would do the same with anything I wrote, so I decided to go over my evaluation summary with her, including the diagnosis. It was one of the most therapeutic encounters I have ever had with a patient with BPD. She was struck not only by the dynamic interpretations of a series of events from her history but also by the fact that she was not alone and crazy and that a class of people existed who shared similar problems. Her first re- sponse after the session was to go to the bookstore, to buy some books on BPD (she found Kemberg especially helpful), and to learn about herself.

Third, diagnosis is essential with some disorders, such as schizophrenia or bipolar disorder, for which treatment decisions depend on accurate di- agnosis. A diagnosis of BPD can be useful as well because the presence of this PD has predictive value vis-his the use of various medications for depression (see, e.g., Gunderson, 1986). Knowing whether a patient who has behaved antisocially actually has the constellation of variables associ- ated with antisocial PD or is acting out a neurotic conflict also is clinically important. A fourth way in which diagnosis is useful is for research purposes, which should ultimately feed back into clinical knowledge. A case in point is the etiology of BPD. For years the only etiological theories were based on reconstructions in psychoanalytic treatment, notably Kernberg’s ( 1975) suggestion that patients with borderline personality organization have a constitutional overabundance of aggressive drive and several less clearly specified hypotheses about pathogenic mother-child interactions in the preoedipal years (e.g., Mahler, Pine, & Bergman, 1975). Once the third edition of the DSM created a PD axis, however, measures began to arise to operationalize diagnoses, notably BPD, which led to research relevant to its etiology. On the basis of this research, clinicians now know that a history of sexual abuse is unusually common in the developmental histories of patients with BPD (Herman, Perry, & Van der Kolk, 1989; Ogata et al., 1990; Westen, Ludolph, Misle, et al., 1990) and that it probably contrib- utes, along with Constitutional factors and the quality of attachment rela- CASE FORMULATION AND DIAGNOSIS 117 & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q tionships in childhood, to the development of the disorder. Without a diagnosis, we might well be continuing to argue about whether sexual abuse plays any role in the disorder, with no basis for choosing among competing theories except for the authority of those who expound them.

Diagnosis has one final utility: It is impossible to do without. Diag- nosis is simply a technical form of categorization, and people cannot nav- igate their way through the world without concepts and categories. Any therapist who claims simply to sit with the patient and experience the person “as he or she is” is cognitively naive. Schemas can lead us astray, but they are essential for thought. Clinicians have to be able to tell whether a patient is psychotic, and they use all kinds of implicit and ex- plicit rules to do so. This is simply how the human brain works. Categories represent observed (or taught) regularities in the world. Without them, every case would be entirely new, and clinicians would have to reinvent a theory of personality for every patient. Even the most stalwart constructiv- ists implicitly assess the way their patients feel, how they handle their feelings, the extent to which they have a good hold on reality, and so forth. This implies assimilation of aspects of patients’ behavior to preex- isting schemas. The question is how to make those schemas clinically useful and more likely to sharpen our understanding of our patients rather than obscure who they are. FUNCTIONAL ASSESSMENT OF PERSONALITY The issue, then, is not whether to give up diagnosis but how to re- frame the diagnostic process so that it is maximally useful. One way to make diagnosis more meaningful when assessing personality is to begin with a functional messment, or functional diagnosis-an assessment of how the individual tends to function cognitively, affectively, and behaviorally under certain conditions relevant to psychological and social adaptation. This means assessing pathology as well as health because helping a patient re- quires knowledge of his or her adaptive capacities as well as areas of dys- function. The kinds of descriptive diagnoses made on Axis I1 can and should be derivative of a functional assessment because once a clini- cian has thoroughly assessed a patient’s personality functioning, a simple prototype-matching procedure can automatically produce a descriptive di- agnosis.

Elsewhere (Westen, 1995, 1996) I have argued that three sets of var- iables, defined by three questions, provide a relatively comprehensive road- map of personality that can guide personality assessment: First, what does the individual wish for, fear, and value and to what extent are these mo- tives conscious and mutually compatible? Second, what are the individual’s 118 DREW WESTEN & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q psychological resources for adapting to internal and external demands?

Third, what is the individual’s capacity for engaging in intimate relation- ships, and how does the individual experience the self, others, and rela- tionships? The interaction of the processes included under these three broad rubrics defines the individual’s personality, as he or she pursues mo- tives and responds to experiences with characteristic ways of thinking, feel- ing, and behaving. From a clinical perspective, a fourth question is devel- opmental: How did these various processes emerge, and at what junctures and in what ways did development go awry?

This view of personality is dynamic and systemic in two senses. First, it views personality as the interaction of psychological processes activated under specific conditions, not as the possession of certain traits to particular degrees. One can, and should, be able to measure each of the variables defined by the model, but assigning people to static categories or ascribing to them relatively static traits is not the goal of assessment because doing so misses the dynamics at the heart of the patient’s personality functioning.

Second, this view does not consider situational demands as being entirely independent of personality characteristics. The environment people expe- rience is not generally independent of their actions (see Wachtel, 1987), and the way they respond is often conditional, depending in part on the presence of certain activating conditions, not automatically and inflexibly elicited regardless of the circumstances (Westen, 199713).

In this section, I flesh out each of these questions and the variables that constitute them (see Exhibit 4.1), again using Mr. D. as an example. Although delineation of these variables drew extensively from research across personality, clinical, and developmental psychology, the first three questions address, respectively, the concerns of classical psychoanalytic the- ory; ego psychology; and object relations theory, relational theories, and self psychology. The fourth cuts across all three. All four questions, al- though theory laden, are close to clinical data and together represent a codification of the kind of assessment most skilled clinicians intuitively make, although the effort here is to systematize those dimensions in a way that is both clinically and empirically sound.

Question 1: What Does the Person Wish for, Fear, and Value, and to what Extent Are These Motives Conscious and Mutually Compatible?

The first question regards motivation: What does the individual wish for, fear, and value? To put it another way, what representations of desired, feared, and valued states has the patient come to associate with a substan- tial enough degree of affect that these representations guide behavior as goal states? These affectively imbued representations can be conscious, un- CASE FORMULATION AND DIAGNOSIS 119 & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q EXHIBIT 4.1 Domains of Personality Functioning in a Comprehensive Assessment I. Motives a. Fears b. Wishes c. Values d. Conflicts among fears, wishes, and values e. Consciousness of dominant motives f. Notable compromise formations 11. Psychological resources a. Cognitive functions 1. Intellectual functioning, verbal and nonverbal skills, memory 2. Cognitive style 3. Coherence or disorder of thought processes 4. Expectancies and belief systems b. Affective experience 1. Intensity of affective experience 2. Variability or lability of affect 3. Tendency to experience positive and negative affect 4. Tendency to experience particular affects 5. Consciousness of affective experience 6. Capacity for experiencing ambivalent emotions c. Affect regulation 1. Conscious coping strategies 2. Defenses 3. Repertoire of affect-regulatory behavior d. Behavioral resources 111. Experience of the self and others and capacity for relatedness a. Cognitive structure of representations of self and others 1. Complexity 2. Differentiation of different representations from each other 3. Integration of diverse elements b. Affect tone of relationship schemas c. Capacity for emotional investment in relationships d. Capacity for investment in values and moral standards e. Understanding of social causality f. Dominant interpersonal concerns: chronically activated interpersonal g. Management of aggressive impulses h. Social skills and interpersonal behavior i. Self-structure wishes, fears, and schemas 1. Sense of self-continuity or coherence; sense of self as a thinker, feeler, 2. Conscious and unconscious representations 3. Self-with-other schemas 4. Self-esteem 5. Feared, wished-for, ought, and ideal self-representations 6. Self-presentation 7. Identity IV. Development a. Developmental level (maturity) of various psychological processes b. Temperament c. Salient developmental experiences and agent; experience of self as being continuous over time 120 DREW WESTEN & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q conscious, or somewhere in between (such as acknowledged but only in alternation, or recognized with considerable clinical probing and support), and they may be congruent or conflicting. They also may be combined to produce compromise formations (see Brenner, 1982), which simultaneously address multiple motives, some or all of which may be unconscious. Recent empirical evidence (summarized by Westen, in press; see Bargh, 1997; McClelland, Koestner, & Weinberger, 1989) supports the long-held clinical view that motives can be either conscious or unconscious, that conscious motives can differ substantially from their unconscious counterparts, and that motives (e-g., the desire to perceive the self accurately and the desire to maximize self-esteem) can conflict and be combined unconsciously to produce compromise responses. A comprehensive assessment of an individual’s motivational structure would, of course, be impossible because it would involve mapping every connection between thought and feeling in the individual’s mind; rather, the major aim of assessment of this domain of personality is to understand the broad sweep of his or her recurring motives, particularly those involved in symptom formation and maintenance: Mr. D. was tom by competing motives to succeed and to fail. On the one hand, he desperately wanted to succeed, in part to match up to a fantasy ideal of his dead father, in part to please his mother and step- father, and in part simply because he had internalized much of the achievement orientation of US. culture. On the other hand, he ap- peared compelled to find ways to fail, and careful examination of his motives in such circumstances led to wishes to be passive and cared for, fears that he would fail that led him to avoid efforts to succeed that seemed to him “doomed,” and conflicts around his masculinity [apparently related in part to his father’s death at a young age] that left him reluctant to take on a more “agentic” role that he associated with manliness [and ultimately death]. Mr. D. was equally conflicted about his relational wishes. On the one hand, he wanted to be closer to people, but he was frightened that he would be rejected and was afraid of his own anger in relationships.

Thus, instead of actually engaging with people, he would often have a running commentary with them in his mind, often filled with ag- gressive content, which served as a compromise among his wishes to connect, his wishes to express his rage at what felt like constant re- jection and humiliation, and his fears of getting involved. Sex was a particular battleground for conflicting motives for Mr. D., who once again wanted to connect emotionally, wanted to satisfy his physical desires, and wanted to feel like a man but was paralyzed by conflicting motives. As noted earlier, he associated female genitals with danger and was both drawn to and repulsed by anal fantasies. [I am not speak- ing theoretically here; he was explicit in describing those fantasies and corresponding fears, with no suggestion necessary on my part.] He also CASE FORh4ULATlON AND DIAGNOSIS 121 & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q had homosexual fantasies that would sometimes intrude during mo- ments of arousal, which was distressing to him.

Question 2: What Psychological Resources (Cognitive, Affective, and Behavioral Dispositions) Does the Individual Have at His or Her Disposal?

The second question includes several subdimensions, largely centering on the individual’s cognitive and affective patterns. The first set of dimen- sions pertains to the cognitive resources at the individual’s disposal. When clinicians assess patients, they try to evaluate several aspects of cognitive functioning. One is intellectual functioning, such as the degree to which the individual can process information efficiently using verbal and visual modes. Contemporary theories of intelligence emphasize multiple intelli- gences and the uses to which people put their intellectual processes in solving problems (Chen & Gardner, 1997; Gardner, 1983; Stemberg, 1985, 1997). Clinical assessment of intelligence similarly entails a functional as- sessment of the way individuals think relative to the tasks that confront them. A dynamically informed cognitive assessment also focuses on the role of motives in channeling intellectual processes, on the extent to which the individual’s affects disrupt them, and in general on the interaction of affect, motivation, and reasoning.

Another cognitive variable is cognitive style (Shapiro, 1965), such as the global, impressionistic, hysterical style that usually co-occurs with de- fenses such as repression, pseudonafvet6, or denial of obvious but unpleas- ant ideas, and the analytical, miss-the-forest-for-the-trees, obsessional style, which is usually accompanied by defenses such as intellectualization and inattention to affect. Similar concepts of cognitive style emerged indepen- dently in other empirical literatures, such as research on field dependence (Berry, 1976), although a dynamic understanding focuses more on cognitive-affective interactions. An additional cognitive variable, first studied by psychoanalytic ego psychologists, is the degree to which an in- dividual’s thought processes are intact or disordered (see Allison, Blatt, & Zimet, 1968; Johnston & Holzman, 1979; Rapaport, Gill, & Schafer, 1945). Finally, cognitive assessment requires attention to cognitive content as well as process, that is, to prominent schemas and beliefs (see, e.g., Beck, 1976, 1993; Weiss & Sampson, 1986). Cognitive therapists pay attention to the more conscious aspects of these schemas, whereas psychodynamic clinicians attend more to unconscious schemas or representations, although the focus of psychodynamic clinicians on such cognitive processes is often obscured by their use of the word fantasies to refer to all manner of psy- chological contents, from the wishful beliefs most people colloquially un- derstand as fantasies, to childhood constructions of reality such as “my father died because I got angry at him.” (To the extent that these cognitive 122 DREW WESTEN & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q contents encode interpersonal information, they attain a salience high- lighted by the third broad question, to be discussed shortly.) Research that distinguishes explicit thought and memory (consciously retrievable and ma- nipulable information) from implicit cognitive processes (such as infor- mation encoded along associational networks, which is inkcessible to con- sciousness; see Holyoak & Spellman, 1993; Schacter, 1992) suggests the need for careful attention to cognitive processes at different levels of con- sciousness because the two systems of thought (implicit and explicit, un- conscious and conscious) are psychologically and neurologically distinct (see Westen, in press).

Another domain of psychological resources is composed of the indi- vidual's chronic affective tendencies. Individuals differ in a number of af- fective dimensions, many of which have been studied empirically, including affective lability (i.e., the extent to which they fluctuate from one emo- tional state to another), affect intensity (i.e., the extent to which emotions are strong; Larsen & Diener, 1987), the extent to which they chronically experience pleasant and unpleasant affects (Buss & Plomin, 1984), the extent to which they experience specific affects such as shame and guilt (see Watson & Clark, 1992; Watson & Tellegen, 1985; Westen, 1994), their comfort with conscious awareness of affect (see Pennebaker, 1989), and, as emphasized by Kemberg (1975, 1984) and subsequently examined empirically (e.g., Baker, Silk, Westen, Nigg, & Lohr, 1992; Sincoff, 1992), their ability to recognize and experience conflicting affective states and appraisals simultaneously (i.e., the capacity for ambivalence).

An important domain of affective functioning that is currently re- ceiving more widespread attention by psychologists is affect regulation, which refers to the conscious and unconscious procedures used to maximize pleasant and minimize unpleasant emotions (see Dozier & Kobak, 1992; Mayer, Salovey, Gomberg-Kaufman, & Blainey, 1991; Westen, 1985, 1994; Westen, Muderrisoglu, Fowler, Shedler, & Koren, 1997). People regulate affects in many ways, using conscious coping strategies (e.g., anticipation, cognitive reframing, self-disnaction, suppression, humor, etc.), unconscious defenses (see Perry & Cooper, 1989; Vaillant, 1992), and behaviors aimed at altering reality to eliminate an aversive situation or at altering the affect directly (e.g., by ingesting drugs or alcohol; Haan, 1977). Finally, assessment of psychological resources includes an assessment of the way the individual chronically tends to behave, which is the domain of personality most often assessed by personality psychologists who endorse the FFM. Although many aspects of behavior are learned, temperament clearly plays an important role in influencing general tendencies, such as activity level, extraversion, and impulse regulation (i.e., the capacity to delay impulse and action). Too often, dynamic therapists assume that if they address motives and conflicts, behavior will change. Sometimes ad- dressing a conflict (e.g., ambivalent feelings about becoming successful) will CASE FORMULATlON AND DIAGNOSIS 123 & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q produce dramatic changes in skills (e.g., study or work skills). At other times, however, even a skill deficit (e.g., problematic social skills) that originally had its origins in conflict (e.g., avoiding the discomfort associated with interacting with a capricious or inept attachment figure) will not remit with dynamic work because it has become a functionally autonomous be- havioral pattern. As Bandura (1977, 1986) has demonstrated, people will not show competent behavior unless they not only have relevant motives and positive expectancies about the likely outcome of producing the be- havior but also the skills to carry it out.

Mr. D., with respect to cognitive resources, was verbally intelligent, although his conflicts about success inhibited him from expressing much of his ability to think with words occupationally. Instead, he worked for several years as a cabinetmaker, where he developed skills that instead made use of his considerable motoric intelligence. His cognitive style was obsessional, in that he would often ruminate on details of conversations, minor mistakes in cabinet construction, and the like. He showed no signs of disordered thinking. A number of cognitive constructions, many of which are addressed by Question 3 [e.g., beliefs about his father’s death, beliefs about his inadequacy, etc.], influenced his mood and behavior. With respect to affective resources, he had access to a range of emotions and was not particularly labile affectively, although his emotions could at times be strong. He tended to experience more negative than positive affect, especially depression and anxiety, and he was often angry but uncomfortable with expressing it. He tended to use obsessional defenses such as intellectualization but also had a number of adaptive coping strategies at his disposal, such as his ability to rely on emotional support from his therapist, parents, and a limited number of close friends. His behavioral style was passive, relatively slow-moving, introverted, and complex with respect to im- pulse regulation because he was often inhibited or reflective before taking action but would sometimes show breakthroughs of poorly reg- ulated impulses [as when he would buy an expensive piece of carpentry equipment with little forethought about how he would pay for it]. Question 3: How Does the Person Experience the Self and Others, and to What Extent Can the Individual Enter Into Intimate Relationships? The third question, regarding interpersonal functioning, forms the content of object relations theory, including self psychology. The term ob- ject relations refers to the cognitive, affective, and motivational processes that underlie the capacity for maintaining intimate relationships. Several dimensions of object relations are empirically distinguishable and have been examined in patients, normal adults, children, and adolescents (see 124 DREW WESTEN & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q Westen, 1991; Westen, Lohr, Silk, et al., 1990; see also Blatt & Lemer, 1983; Stricker & Healey, 1990). These dimensions include the cognitive structure of representations of people (complexity, differentiation, and in- tegration), the affect tone of relationship schemas (i.e., the extent to which the individual expects relationships to be destructive or enriching), the capacity for emotional investment in relationships (i.e., the transcendence of a need-gratifying orientation to others), moral development, the capacity for understanding why people do what they do, the ability to regulate and appropriately express aggression, the dominant interpersonal concerns (fears, wishes, and cognitive constructions) that repetitively emerge in the individual’s relationships and are manifest in narratives of interpersonal encounters (in psychotherapy hours or on projective tests), and social skills.

Another set of variables involves aspects of self (for a more detailed delin- eation, see Westen, 1992; Westen & Cohen, 1992), including the coher- ence of the individual’s sense of self; the nature of chronically recurring self-representations or self-schemas; self-esteem; feared, wished-for, and ideal self-representations that serve as standards or guides for behavior (see Higgins, 1990; Strauman, Lemieux, & Coe, 1993); and what Erikson (1963) referred to as identity, which includes both the sense of self, rep- resentations of self, the recognition of one’s selfhood by the social milieu, and an emotional weighting of elements of self (e.g., roles) the individual experiences as self-defining. Mr. D., like many patients, demonstrated a much more complex, mul- tifaceted picture with respect to his object relations than is often afforded by unidimensional stage theories that place an individual at a single de- velopmental level, such as fixated at a hypothesized rapprochement stage (see Westen, 1989, 1990): Mr. D.’s representations of people were often fairly complex; he rarely showed signs of splitting, was generally clear about whose feelings were whose, and was able to recognize people’s strengths as well as their feelings, even when he was angry or upset. The affective quality of his representations of others was mixed. He did not expect malevolence from others and was able to experience pleasure in the small number of close friendships he had developed over the years, but his experi- ences with people had left him with somewhat negative views of what he could expect to happen in relationships [which were not altogether unrealistic given his experiences and anxieties in interacting with them]. He wanted to connect with people, was able to care about others, and had a well-developed, if sometimes harsh, superego. He also had a relatively solid understanding of why people do what they do, although at times he had trouble using this knowledge in concrete social situations. He had difficulty with regulation of aggression and would frequently be passive or self-punitive rather than appropriately assertive, which contributed to a lingering hostile fantasy life and a tendency at times to behave passive aggressively. His dominant inter- CASE FORMUJATION AND DIAGNOSIS 125 & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q personal concerns centered around rejection, shame, and aloneness. He had the skills to interact with people, but he was frequently inhibited by his anxiety from doing so and would often find himself tongue-tied, especially with women. He had no difficulties experiencing himself as coherent (i-e., he had no dissociative tendencies), but he chronically felt inadequate and perceived himself as failing to meet his own and his parents’ standards for achievement. He often voiced identity con- cerns directly, wondering what he was going to do with his life, where he would fit in, and feeling adrift without either meaningful work or love relationships that were sustaining.

Question 4: How Do These Processes Develop?

The final question is one of development. A distinguishing feature of psychodynamic theory has always been the assumption that to understand any adult phenomenon one must understand its genesis. Although theorists have often erred in assuming that all pathological variants represent fixa- tion or regression points along normal developmental continua, they have emphasized that to understand why an individual is behaving in a way that seems maladaptive or unintelligible, one should ask several questions: Does this behavior (e.g., throwing tantrums in an older child, ending relation- ships precipitously when they are momentarily ungratify ing in an adult) resemble the behavior of younger children? What events transpired during development that made this individual vulnerable? When did the individ- ual first behave this way, and what problem was this behavior an attempt to solve? Underlying these questions is the view that people often find themselves saddled as adults with ways of seeking pleasure, regulating un- pleasant affects, and construing events that were forged in childhood. Be- cause these thoughts, motives, and affect regulation strategies became au- tomatic or unconscious through repeated use, and because consciousness of some of them may have produced discomfort, they may not have been appropriately revised in light of adult circumstances and cognition. Thus, case formulation requires attention to ways in which people have maintained response tendencies that are no longer adaptive (or may never have been adaptive). A thorough case formulation also attends to two other developmental questions. First, what are the temperamental con- tributions to enduring personality dispositions, and how are constitutional factors implicated in the etiology of the patient’s symptoms? Research in behavioral genetics has shown that many aspects of personality, such as negative affectivity and introversion-extraversion, are highly heritable (see, e.g., Kagan 6r Snidman, 1991; Plomin, Chipuer, 6.1 Loehlin, 1990; Tellegen, Lykken, Bouchard, Wilcox, & Rich, 1988). A second develop- mental dimension involves salient developmental experiences, such as pa- rental behavior, parental loss, neglect, and abuse. A growing body of re- I26 DREW WESTEN & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q search on attachment establishes the role of attachment patterns, as early as the first 12 months of life, in shaping later social adjustment (Bretherton, 1990; Lyons-Ruth, Connell, Grunebaum, & Botein, 1990; Sroufe & Flee- son, 1986). Adolescent drug use and abuse, as well as personality charac- teristics such as rigidity or manipulativeness, can be predicted from moth- ers’ interactions with their preschool children (Shedler & Block, 1990). Research has similarly documented links between experiences such as abuse, neglect, and disruptions in caretaking throughout childhood with later interpersonal pathology and personality disorders (see Tizard & Hodges, 1978; Westen, Ludolph, Block, et al., 1990). Understanding a given individual’s personality requires examination of the way temperament and life experiences interact to shape his or her development in the domains outlined under the first three questions (i.e., motives, psychological resources, and experience of self and others). For example, an event such as sexual abuse may have a differential impact on various psychological processes. In one individual it may lead to malevolent expectations of relationships but not to global alterations in cognitive func- tioning. In another, it may lead to a disruption in verbal intelligence caused by a defensive shutting down of semantic knowledge as a way of constrict- ing associative memory- Mr. D.’s history suggests likely contributions of both biological and experiential events to his psychopathology. His mother was chronically depressed, which suggests a biological diathesis, a psychosocial diath- esis, or both. In fact, depression ran throughout her family, even in cousins with whom she had minimal contact, suggesting a biological contribution. Given the high comorbidity rates of depressive and anx- iety disorders, Mr. D.’s anxiety-and his tendency to experience neg- ative affective states more generally-probably had origins in part in temperament. This temperamental vulnerability was likely exacerbated by experiences that could lead to distress and pathogenic representa- tions of self and relationships in even a temperamentally more resilient child, such as his mother’s chronic depression and his father’s violent death. Given the age at which he experienced both (the former from as early as he could remember) and the immaturity and egocentrism of much preschool thought, he was likely to have constructed an un- derstanding of the causes of the misery in his family that put his own actions and fantasies at center stage. Precisely how and why he devel- oped such negative self-representations and social anxiety is unclear, although tendencies of this sort can be self-reinforcing. As an adult, for example, he developed negative expectancies and a conditioned anxiety response to sexual situations after his first sexual encounters, which involved erectile failures. This only increased his anxiety, which appeared to have been spurred initially by unbidden images and anx- iety-filled fantasies and associations to sexuality. CASE FORMULATION AND DIAGNOSIS 127 & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q FROM FUNCTIONAL DIAGNOSIS TO DESCRIPTIVE DIAGNOSIS The material in italics describing Mr. D. throughout this chapter, as scant as it is, probably seems to most readers considerably richer and more descriptive than his DSM-N diagnosis, and, at first glance, the two appear unrelated. In fact, however, a descriptive diagnosis can be derived from a functional diagnosis, so that the advantages of the former (e.g., commu- nication among clinicians, research on etiology, etc.) can be garnered with- out the limitations. I conclude by briefly describing a procedure by which such a translation can be made for both empirical and clinical purposes.

Jonathan Shedler and I have spent the past several years developing a Q-sort instrument for assessing personality dynamics and pathology called the Shedler-Westen Assessment Procedure (SWAP-200; Shedler & Wes- ten, 1996). A Q-sort is a scaling procedure that has been receiving in- creased use in personality research because of its ability to translate clinical or quasi-clinical judgments into quantified narrative statements (see Block, 1978). As in other Q-sorts, the statements that make up the SWAP-200 Q-sort are written in clear language with minimal jargon (to avoid multiple interpretations of the construct being assessed) and are printed on note- cards, one per card. The task for the observer (a clinician, interviewer, or coder who has watched an interview) is to sort the cards into eight piles, numbered 0-7, to indicate the degree to which each statement describes the patient. Thus, items in Pile 7 are absolutely defining of the individual’s personality, whereas those in Pile 0 do not describe the individual at all.

This procedure yields a narrative description of the patient in de- scending order of diagnosticity (from Pile 7 downward), as well as a quan- tified assessment of the patient’s personality, because each statement now has a number attached to it, from 0 to 7. Although the statements in the SWAP-200 were not designed explicitly to assess the dimensions of per- sonality delineated here, we did attempt to cover these domains. Initial validity and reliability studies have been promising (see Shedler & Westen, 1996), and we have recently completed a study of 797 patients with PDs who were described by a random national sample of psychologists and psy- chiatrists from the registers of the American Psychiatric Association and the American Psychological Association, which has allowed us to examine empirically the statements that appear best to describe patients with each of the PDs listed in the DSM-IV to ascertain using cluster analytic meth- ods whether alternative ways of categorizing patients with PDs might be more useful, and examine the relation between various genetic and devel- opmental history variables (e.g., familial alcoholism, abuse, loss; Question 4) and personality characteristics (Questions 1-3; Westen & Shedler, 1998a, 1998b). The SWAP-200 can be used to provide a quantified nar- rative description of the personality characteristics of any patient, provided that the observer (such as the patient’s therapist) knows the patient well 128 DREW WESTEN & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q enough to do so (which we have operationally defined as a minimum of five sessions’ contact).

Consider the statements that would receive highest placement (Piles 6 and 7) in the sort for Mr. D. (in no particular order): m m m m m m m I m m Has difficulty acknowledging or expressing anger.

Tends to be anxious.

Tends to feel helpless; feels own wishes or actions have little effect. Is empathic; is sensitive and responsive to other peoples’ needs and feelings.

Tends to feel ashamed or embarrassed. Is articulate; can express self well in words.

Tends to be shy or reserved in social situations. Is capable of hearing information that is emotionally threat- ening (i.e., that challenges cherished beliefs, perceptions, and self-perceptions) and can use and benefit from it. Appears to associate sexual activity with danger (e.g., injury, punishment, contamination, etc.), whether consciously or unconsciously.

Has trouble making decisions; tends to be indecisive or to vacillate when faced with choices.

Experiences a specific sexual dysfunction during sexual inter- course or attempts at intercourse (e.g., inhibited orgasm or vaginismus in women, impotence or premature ejaculation in men).

Tends to be inhibited or constricted; has difficulty allowing self to acknowledge or express wishes and impulses.

Tends to think in abstract and intellectualized terms, even in matters of personal import.

Tends to avoid social situations because of fear of embarrass- ment or humiliation.

Tends to be passive and unassertive. Is psychologically insightful; is able to understand self and others in subtle and sophisticated ways.

Tends to feel unhappy, depressed, or despondent.

The following items would receive moderately high placement (e.g., Piles 4 and 5) but would not be in the most diagnostic piles: D Has moral and ethical standards and strives to live up to them. m Has the capacity to recognize alternative viewpoints, even in matters that stir up strong feelings. Tends to be self-critical; seems intolerant of own human de- fects. CASE FORMULATION AND DIAGNOSIS 129 & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q Tends to fear rejection or abandonment by those who are emotionally significant. Tends to be critical of others. Is able to form close and lasting friendships characterized by mutual support and sharing of experiences. m Is unsure whether he or she is heterosexual, homosexual, or bisexual.

Lacks a stable image of who he or she is or would like to become (e.g., attitudes, values, and goals may be unstable and changing). Tends to be ingratiating or submissive (e.g., may consent to things he or she does not agree with or does not want to do in the hope of getting support or approval). B Tends to be conflicted about authority (e.g., may feel he or she must submit, rebel against, win over, defeat, etc.). m Tends to act impulsively, without regard for consequences. m Tends to abuse alcohol.

Tends to feel like an outcast or outsider; feels as if he or she does not truly belong. Tends to oscillate between undercontrol and overcontrol of needs and impulses (i.e., needs and wishes are expressed im- pulsively and with little regard for consequences, or else dis- avowed and permitted virtually no expression). I Is creative; is able to see things and approach problems in novel ways. I Tends to see normal sexual experiences as somehow revolting or disgusting. Appreciates and responds to humor. Tends to be angry or hostile (whether consciously or uncon- sciously).

Tends to be overly needy or dependent; requires excessive reassurance or approval.

Several aspects of this characterization of Mr. D. are noteworthy. First, the procedure offers a rich portrait of the patient’s personality dynamics and psychopathology. Any competent clinician reading these statements would have a sense of what Mr. D. is like as a person and what issues call for therapeutic attention. Second, the virtue of the Q-sort method is that it turns clinically meaningful statements into quantifiable items on which two observers can agree. We have found, for example, that the interrater reliability for the combined scores of two coders observing the same loosely structured clinical interview on a given patient is .75 and that this profile of statement scores correlates, on the average, with the treating clinician’s Q-sort description of the patient (r = S4). What the Q-sort procedure I30 DREW WESTEN & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q essentially allows clinicians to do is to make a case formulation using a common language, so that any two well-trained clinicians should be able to provide case formulations on a patient that are not only qualitatively but quantitatively highly similar.

Third, the SWAP-200 description of Mr. D., unlike his DSM-N diagnosis, provides a portrait of both his strengths and weaknesses, which allows the clinician to know what needs to be done, what does not, and what strengths can be called on in the service of treatment. Fourth, it provides a clearer sense of diagnosis in patients such as Mr. D. who have clear personality pathology but not a PD. The high placement of health items-such as Mr. D.’s capacity for empathy or for making use of infor- mation that is threatening but potentially useful-makes clear that Mr. D. may be depressed, anxious, and sexually conflicted but that he does not have a PD despite areas of maladaptation.

Fifth, this system does not define a patient’s diagnosis in terms of the presence or absence of a small number of symptoms (e.g., four of seven symptoms of avoidant PD). Rather, it provides a broad profile of attributes, ranging from those that are extremely true of the patient, through those that are moderately true, through those that are not true at all. In so doing, it allows the clinician to make 200 discriminations about the patient (be- cause the Q-sort has 200 items) rather than trying to decide whether he or she meets a small number of criteria for a disorder that may itself be somewhat arbitrarily defined. The Q-sort procedure also does not require that clinicians make dichotomous (present or absent) choices about vari- ables (e.g., social anxiety, depression, rejection sensitivity, etc.) that are continuous in nature. Particularly useful about this method of diagnosis is that it allows for more sophisticated research about treatment and treat- ment outcome. Not only can researchers use this instrument to determine precisely which aspects of personality change with which kind of treatment after what average duration, but they also can use it to construct scales- which are nothing more than lists of statements-that can indicate to clinicians which patients might benefit from particular kinds of interven- tions. For example, by examining a large sample of depressed patients who responded to selective serotonin reuptake inhibitors (SSRIs) and compar- ing them with a sample that did not, one can develop a profile of the personality characteristics of patients whose symptoms are most likely to “listen to Prozac.” This would have been useful for Mr. D., who, in fact, responded well to treatment with SSRIs. A final aspect of this Q-sort description is particularly important from the perspective of the present chapter: It can be translated directly into a descriptive diagnosis by a process of prototype matching. For research pur- poses, one can correlate a patient’s Q-sort profile with the obtained aggre- gated profile of a large sample of patients with various disorders to deter- mine the extent to which the patient resembles patients with each disorder.

CASE FORMULATION AND DIAGNOSIS 131 & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q Thus, if Mr. D.’s profile correlates highly with the average profile of a group of 40 patients with depressive PD but only moderately with that of a sample of avoidant patients, one can describe his pathology with confidence in terms of the degree of match, and with precision in terms of the size of the correlation. In our cluster analytic study, we found a category of “neu- rotic depressives” with which Mr. D.’s profile would correlate highly. For clinical purposes, clinicians can intuitively perform the same operation on the basis of their experience by comparing the profile of statements that are descriptive of their patient with other patients they have worked with to arrive at an Axis I1 diagnosis. The advantage of this method clinically is that the diagnosis flows directly from the functional assessment. CONCLUSION In recent years, the gap between clinicians and researchers has been increasing, particularly with respect to the diagnosis of personality (Westen, 1997a). The elimination of neurotic character patterns from official psy- chiatric discourse, the development of structured interviews that diverge from clinical practice, the attempt to describe personality patterns without reference to personality theories, and a variety of other trends have led many clinicians to consider the official nosology and research using it ir- relevant to clinical practice.

We need not eliminate Axis I1 in favor of a five-factor model of personality based on self-repom. And we need not, and should not, ghet- toize some of the most important aspects of personality and its dynamics, such as defenses, to appendixes in the various editions of the DSM, leaving an eviscerated official diagnostic system. These options will not work be- cause they eliminate from the assessment of personality precisely what, from a clinical perspective, personality is all about: the tendency to respond cognitively, affectively, and behaviorally under certain specifiable circum- stances in ways that define both the individual’s enduring dispositions and aspects of his or her character that he or she could profitably change. REFERENCES Allison, J., Blatt, S., & Zimet, C. (1968). The interpretation of psychobgicd tests. New York: Harper & Row. American Psychiatric Association. (1994). Diagnostic and statistical munwl ofmend dismders (4th ed.). Washington, DC: Author.

Baker, L., Silk, K. R., Westen, D., Nigg, J. T., & Lohr, N. E. (1992). Malevolence, splitting, and parental ratings by borderlines. Journal of Nervous and Mend hease, 180, 258-264. 132 DREW WESTEN & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: Prentice Hall.

Bandura, A. (1986). Sod foundations of thought and action. Englewood Cliffs, NJ:

Bargh, J- (1997). The automaticity of everyday life. In J. S. Wyer, J. (Ed.), Ad- vances in social cognition (Vol. 10). Hillsdale, NJ: Erlbaum.

Beck, A. T. (1976). Cognitive therapy and the emotional disurders. Madison, CT: International Universities Press.

Beck, A. T. (1993). Cognitive therapy: Past, present, and future. Journal of Con- sulting and Clinical Psychology, 61, 194-198.

Bell, E., & Jackson, D. (1992). The structure of personality disorders in DSM-111. Acta Psychiuhica Scandinuvku, 85, 279-287.

Berry, J. W. (1976). Human ecology and cognitive style: Comparative studies in cultural and psychobgud adaptation. New York: Sage. Blatt, S. J., & Lemer, H. (1983). Investigations in the psychoanalytic theory of object relations and object representations. In J. Masling (Ed.), Ernpirid stud- ies of psychoudytic theuries (Vol. 1, pp. 189-249). Hillsdale, NJ: Erlbaum.

Block, J. (1978). The Q-surt method in personality assessment and psychiatric research. Palo Alto, CA: Consulting Psychologists Press.

Brenner, C. (1982). The mind in conflict. Madison, CT: International Universities Press. Bretherton, I. (1990). Communication patterns, internal working models, and the intergenerational transmission of attachment relationships. Infant Mend Health Journal, 11, 237-252. Buss, A., & Plomin, R. (1984). Temperament: Early deueloping personality muits. Hillsdale, NJ: Erlbaum.

Chen, J.-Q., & Gardner, H. (1997). Alternative assessment from a multiple in- telligences theoretical perspective. In D. P. Flanagan, J. L. Genshaft, & P. L. Harrison (Eds.), Contemporay intellectual assessment: Theories, tests, and issues (pp. 105-121). New York: Guilford Press. Dozier, M., & Kobak, R. (1992). Psychophysiology in attachment interviews: Con- verging evidence for deactivating strategies. Child Development, 63, 1473 - 1480.

Prentice Hall.

Erikson, E. (1963). Childhood and society. New York: Norton.

Funder, D., & Colvin, C. R. (1991). Explorations in behavioral consistency: Prop- erties of persons, situations, and behaviors. Joud of Personality and Social Psychology, 60, 773-794.

Gardner, H. (1983). Frames of mind: The theory of multiple intelligences. New York:

Goldberg, L. (1993). The structure of phenotypic personality traits. American Psy- Green, M. A., & Curtis, G. C. (1988). Personality disorders in panic patients: Response to termination of antipanic medication. Journal of Personality Dis- Basic Books. chologist, 48, 26-34. order~, 2,303-314. CASE FORMULATION AND DIAGNOSIS 133 & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q Gunderson, J. G. (1986). Pharmacotherapy for patients with borderline personality Haan, N. (1977). Coping and defending: Processes of self-environment organization.

Herman, J., Perry, J. C., & Van der Kolk, B. A. (1989). Childhood trauma in borderline personality disorder. American Jouwl of Psychiatry, 146,490-495.

Higgins, E. T. ( 1990). Personality, social psychology, and person-situation relations:

Standards and knowledge activation as a common language. In L. Pervin (Ed.), Handbook of persodty: Theory and research (pp. 301-338). New York:

Guilford Press.

Holyoak, K., & Spellman, B. (1993). Thinking. Ann& Review of Psychology, 44, 265-315.

Hooley, J. M. (1987). The nature and origins of expressed emotion. In K. Hahlweg & M. J. Goldstein (Eds.), Understunding mujor mental disorder: The conmibution of family interaction research (pp. 176-194). New York: Family Process Press.

John, 0. (1990). The “big five” factor taxonomy: Dimensions of personality in the natural language and in questionnaires. In L. Pervin (Ed.), Handbook of per- sonality: Themy and research (pp. 66-100). New York: Guilford Press.

Johnston, M. H., & Holzman, P. (1979). Assessing schizophrenic thinking: A clinical and research instrumt for measuring thought disorder. San Francisco: Jossey- Bass.

Kagan, J., & Snidman, N. (1991). Temperamental factors in human development. American Psychologist, 46, 856-862.

Kenrick, D., & Funder, D. (1988). Profiting from controversy: Lessons from the person-situation debate. American Psychologist, 43, 23-34.

Kemberg, 0. (1975). Borderline conditions and pathologrcal narcissism. Northvale, NJ: Jason Aronson.

Kemberg, 0. (1984). Severe personality disorders. New Haven, CT: Yale University Press.

Larsen, R., & Diener, E. (1987). Affect intensity as an individual difference char- acteristic: A review. Journal of Reseurch in Personality, 21, 1-39. Lyons-Ruth, K., Connell, D., Grunebaum, H., & Botein, S. (1990). Infants at social risk: Maternal depression and family support services as mediators of infant development and security of attachment. Child Development, 61, 85- 98. Mahler, M., Pine, E, & Bergman, A. (1975). The psychological birch of the human infant: Symbiosis and individuation. New York: Basic Books.

Mayer, J., Salovey, P., Gomberg-Kaufman, S., & Blainey, K. (1991). A broader conception of mood experience. Journal of Personality and Sod Psychology, 60, 100-111.

McClelland, D. C., Koestner, R., & Weinberger, J. (1989). How do self-attributed and implicit motives differ? Psychological Review, 96, 690-702.

McCrae, R., & Costa, P. (1990). Personality in adulthood. New York: Guilford Press.

disorder.

Archives of Gem& Psychiutry, 43, 698-700.

New York: Academic Press. 134 DREW WESTEN & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q Miklowitz, D., Velligan, D., Goldstein, M. J., Neuchterlein, K., Gitlin, M., Ranlett, G., & Doane, J- (1991). Communication deviance in families of schizophrenic and manic patients. Journal of Abnd Psychology, 100, 163-173.

Mischel, W. (1968). Personality and assessment. New York: Wiley. Morey, L. (1988). Personality disorders in DSM-111 and DSM-111-R: Convergence, coverage, and internal consistency. American Journal of Psychiutry, 145, 573- 577. Ogata, S., Silk, K. R., Goodrich, S., Lohr, N. E., Westen, D., & Hill, E. (1990).

Childhood abuse and clinical symptoms in borderline personality disorder. American Journal of Psychiatry, 147, 1008-1013.

Oldham, J., Skodol, A., Kellman, H., Hyler, S., Rosnick, L., & Davies, M. (1992).

Diagnosis of DSM-111-R personality disorders by two semistructured inter- views: Patterns of comorbidity. American Journal of Psychtry, 149, 213-220.

Pennebaker, J. (1989). Stream of consciousness and stress: Levels of thinking. In J. S. Uleman & J. A. Bargh (Eds.), Unintended thought (pp. 327-350). New York: Guilford Press.

Perry, J. C., & Cooper, S. H. (1989). Empirical studies of defense mechanisms: I. Clinical interview and life vignette ratings. Archives of General Psychiany, 46, Plomin, R., Chipuer, H., & Loehlin, J. C. (1990). Behavioral genetics and per- sonality. In L. Pervin (Ed.), Handbook of personality: Theory and research (pp. 225-243). New York: Guilford Press.

Rapaport, D., Gill, M., & Schafer, R. (1945). Diagnostic psychologid testing (Vol. 1). Chicago: Yearbook Publishers.

Schacter, D. L. (1992). Understanding implicit memory: A cognitive neuroscience approach. American Psychologist, 47, 559-569.

Shapiro, D. (1965). Neurotic styles. New York: Basic Books. Shea, M., Glass, D., Pilkonis, P., Watkins, J., & Docherty, J. (1987). Frequency and implications of personality disorders in a sample of depressed outpatients. Journal of Personality Disorders, 1, 27-42.

Shedler, J., & Block, J. (1990). Adolescent drug use and psychological health: A longitudinal inquiry. American Psychologist, 45, 612-630.

Shedler, J., & Westen, D. (1996). Refining Axis II ad its measurement: A Q-sm procedure for assessing personality disorders and dynamics. Unpublished manu- script, Harvard Medical School, New Haven, CT. Sincoff, J. (1992). Ambivalence and defense: Effects of a repressive style of normal adolescents’ and young adults’ mixed feelings. Journal of Abnd Psychology, 101, 251-256.

Sroufe, L. A., & Fleeson, W. (1986). Attachment and the construction of rela- tionships. In W. W. Hartup & Z. Rubin (Eds.), Relationships and development. Hillsdale, NJ: Erlbaum.

Stemberg, R. (1985). Human intelligence: The model is the message. Science, 4730, 1 11 1-1 1 17.

444-460. CASE FORMULATION AND DIAGNOSIS 135 & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q Stemberg, R. J. (1997). The triarchic theory of intelligence. In D. P. Flanagan, J. L. Genshaft, & P. L. Harrison (Eds.), Contan~orary intellectual assessment: Theories, tests, and issues (pp. 92-104). New York: Guilford Press.

Strauman, T., Lemieux, A., & Coe, C. (1993). Self-discrepancy and natural killer cell activity: Immunological consequences of negative self-evaluation. Journal of Personality and Sockd Psychobgy, 64, 1042-1052.

Stricker, G., & Healey, B. J. (1990). Projective assessment of object relations: A review of the empirical literature. Psychological Assessment, 2, 2 19-230.

Tellegen, A., Lykken, D. T., Bouchard, T. J., Jr., Wilcox, K. J-, & Rich, S. (1988).

Personality similarity in twins reared apart and together. Journal of Personality Tizard, B., & Hodges, J. (1978). The effects of early institutional rearing on the development of eight-year-old children. Journal of Child Psychology and Psy- Vaillant, G. (Ed.). (1992). Ego mechanisms of defense: A gude for clinicians and researchers. Washington, DC: American Psychiatric Press.

Wachtel, P. L. (1987). Action and insight. New York: Guilford Press.

Watson, D., & Clark, L. A. (1992). Affects separable and inseparable: On the hierarchical arrangement of the negative affects. Journal of Personality and Sod Psychology, 62, 489-505.

Watson, D., & Tellegen, A. (1985). Toward a consensual structure of mood. Psy- chological Bulletin, 98, 219-225.

Weiss, J., & Sampson, H. (1986). Testing alternative psychoanalytic explanations of the therapeutic process. In J. Masling (Ed.), Empirical studies of psychn- dytic theories (Vol. 2, pp. 1-26). Hillsdale, NJ: Analytic Press.

Westen, D. (1985). Self and society: Narcissism, collectivism, and the development of morals. Cambridge, England: Cambridge University Press.

Westen, D. (1989). Are “primitive” object relations really preoedipal? American Journal of Orthopsyc)uatry, 59, 331 -345.

Westen, D. (1990). Toward a revised theory of borderline object relations: Impli- cations of empirical research. Inmtional Journal of Psycho-Analysis, 71, 66 1 -693.

Westen, D. (1991). Social cognition and object relations. Psychological Bulletin, 109, 429-455.

Westen, D. (1992). The cognitive self and the psychoanalytic self: Can we put our selves together? Psychological Inquiry, 3, 1-13.

Westen, D. (1994). Toward an integrative model of affect regulation: Applications to social-psychological research. Journal of Personality, 62, 641 -647.

Westen, D. (1995). A clinical-empirical model of personality: Life after the Mis- chelian ice age and the NEO-lithic era. Journal of Personality, 63, 495-524.

Westen, D. (1996). A model and a method for uncovering the nomothetic from the idiographic: A comparison with the Five Factor Model. Jod of Research in Personality, 30, 400-413. and SO& P~y~holog~, 54, 1031-1039. chotherapy, 19, 99-108. I36 DREW WESTEN & R S \ U L J K W $ P H U L F D Q 3 V \ F K R O R J L F D O $ V V R F L D W L R Q 1 R W I R U I X U W K H U G L V W U L E X W L R Q Westen, D. (1997a). Divergences between clinical and research methods for as- sessing personality disorders: Implications for research and the evolution of Axis 11. American Journal of Psychtry, 154, 895-903.

Westen, D. (1997b). Toward an empirically and clinically sound theory of moti- vation. International Journal of Psycho-Analysis, 78, 521-548.

Westen, D. (in press). The scientific legacy of Sigmund Freud: Toward a psycho- dynamically informed psychological science. Psychological Bulletin. Westen, D., & Arkowitz, L. (1998). Most patients treated in clinical practice for personality pnthobgy cannot be diugnosed on Axls II. Unpublished manuscript, Harvard Medical School, Cambridge, MA.

Westen, D., 6r Cohen, R. (1992). “The self” in borderline personality disorder. In 2. Segal & S. Blatt (Eds.), The self in emotional distress: Cognitive and psycho- dynumic perspectives (pp. 334-360). New York: Guilford Press.

Westen, D., Lohr, N., Silk, K., Gold, L., & Kerber, K. (1990). Object relations and social cognition in borderlines, major depressives, and normals: A TAT analysis. Psychologxd Assessment: A Journal of Consulting and Clinical Psychol- Westen, D., Ludolph, P., Block, J., Wixom, J., & Wiss, E C. (1990). Develop- mental history and object relations in psychiatrically disturbed adolescent girls. American Journal of Psychiatry, 147, 1061-1068.

Westen, D., Ludolph, P., Misle, B., Ruffins, S., & Block, M. J. (1990). Physical and sexual abuse in adolescent girls with borderline personality disorder. American Journal of orthopsychiutry, 60, 55-66.

Westen, D., Muderrisoglu, S., Fowler, C., Shedler, J., & Koren, D. (1997). Affect regulation and affective experience: Individual differences, group differences, and measurement using a Q-sort procedure. Joud of consulting and Clinical Psychology, 65, 429-439.

Westen, D., & Shedler, J. (1998a). Revising and assessing Axis II: I. Developing a clinically and empirically valid method. Unpublished manuscript, Harvard Med- ical School, Cambridge, MA.

Westen, D., & Shedler, J. (1998b). Revising and assessing Axis II. II. Tward an empirically and clinically sensible taxonomy of personality disorders. Unpublished manuscript, Harvard Medical School, Cambridge, MA.

Widiger, T., & Frances, A. (1994). Toward a dimensional model for the personality disorders. In P. Costa & T. Widiger (Eds.), Personality disorders and the five- fracm model of Personality (pp. 19-39). Washington, DC: American Psycho- logical Association. ogy, 2,355-364. CASE FORMULATION AND DIAGNOSIS 137