Naomi E. Lohr

Object relations and social cognition in borderlines, major depressives, and normals : a thematic apperception test analysis

This study compared reliably diagnosed borderline personality disorder patients (n = 35) with major depressives (n = 25) and normals (n = 30) on 4 dimensions of object relations and social cognition coded from Thematic Apperception Test (TAT) responses: Complexity of Representations of People, Affect-Tone of Relationship Paradigms (malevolent to benevolent), Capacity for Emotional Investment in Relationships, and Understanding of Social Causality. As predicted, borderlines scored significantly lower on all 4 scales than did normals and lower on Affect-Tone and Capacity for Emotional Investment than did nonborderline major depressives. Borderlines also produced more pathological responses than did both groups on every scale, indicating more poorly differentiated representations, grossly illogical attributions, malevolent expectations, and need-gratifying relationship paradigms. The results suggest the importance of distinguishing several interdependent but distinct cognitive-affective dimensions of object relations and the potential utility of assessing object relations and social cognition from TAT responses. The psychological processes underlying the interpersonal pathology of patients with borderline personality disorder (BPD), that is, their distorted object relations, have been conceptualized in various ways by different theorists. Most argue that a disturbance in the first 3 years of life leads to the continued use of developmentally primitive modes of relating in adulthood (Kernberg, 1975; Masterson, 1976). Although theorists and clinicians (see Greenberg & Mitchell, 1983) often speak of levels of object relations as unitary phenomena, from developmentally immature to mature, the term object relations refers to a congeries of cognitive and affective functions and structures, including ways of representing people and relationships, rules of inference for interpreting the causes of peoples feelings, behaviors, interpersonal wishes, conflicts, and so forth. The concept of general levels of object relations is clinically an indispensable heuristic, but these levels should be understood as being composed of several interdependent but distinct developmental lines that differ in their maturity and quality among individuals as well as within a single individual at any given time (Westen, 1989, 1990, in press-b). The aim of the present study is to explore the nature of different dimensions of the cognitive-af

Malevolent Object Representations in Borderline Personality Disorder and Major Depression

To study malevolent representations, earliest memories were reliably coded on scales of affect tone. Ss were diagnosed with borderline personality disorder: 31 without and 30 with concurrent major depression. Nonborderline comparison subjects had either major depressive disorder (n = 26) or no psychiatric diagnosis (n = 30). Borderline subjects were discriminated from comparison subjects by their more malevolent representations; they more frequently produced memories involving deliberate injury; and they portrayed potential helpers as less helpful. Results suggest the diagnostic significance of malevolent representations, which need to be explained by any theory of borderline personality disorder.

Malevolence, splitting, and parental ratings by borderlines

Malevolent object relations as well as splitting have long been considered by psychodynamic theorists as central features of borderline personality disorder. We tested the hypotheses that borderlines would a) perceive their parents more negatively than both nonborderline major depressive patients and nonpatient normal controls, and b) split their representations of their parents into opposites more than the comparison subjects. Borderlines (N=31), who were identified by the Diagnostic Interview for Borderlines, Research Diagnostic Criteria major depressives (N=15), and nonpatient controls (N=14) were asked to rate each parent on the Adjective Check List (ACL; Gough and Heilbrun, 1983). Seven ACL scales were studied: Favorable, Unfavorable, Critical Parent, Nurturing Parent, Nurturance, Aggression, and Dominance. Correlations were performed between scores for mother and father on the various scales for each of the three cohorts. Analysis of variance and one-way Wests with Bonferroni correction were used to test group differences. Borderlines rated their parents, especially their fathers, not only as more unfavorable on negative scales than depressives or normals, but as less favorable on positive scales than the comparison groups. Analysis of covariance revealed that a significant portion of the variance in father scores, but not in mother scores, was related to age of respondent and history of sexual abuse. While borderlines did not appear to split their parents into one good and one bad parent, they did show significantly less correlation between parents on the Favorable scale when compared with either depressives or normal subjects. The results imply that borderlines have a greater tendency to view the world in negative, malevolent ways than to split their object representations.

Object Relations in Borderlines, Depressives, and Normals: An Examination of Human Responses on the Rorschach

Recently, researchers and clinicians have become increasingly interested in diagnostic distinctions between borderline and mood disorders. Object relations theory provides a useful framework for the comparison of these two overlapping diagnostic categories. In our study, a measure of object relations as represented on the Rorschach, developed by Blatt, Brenneis, Schimeck, and Glick (1976), was applied to data produced by borderline and depressive inpatients and by normal comparison subjects. Portions of the Blatt measure that tap the subjects experience of human action and interaction distinguish among the three diagnostic groups. Specifically, borderlines tend to understand human action as more highly motivated and human interaction as more malevolent in nature than do either depressive or normals. The data indicate that borderlines experience the object-relational world in a way that is fundamentally different from the way normals and depressives perceive it. Implications are discussed for theories of borderline object relations.

History of childhood sexual abuse and general psychopathology

Reported history of childhood sexual abuse (CSA) was correlated with general measures of psychopathology on the SCL-90-R in a sample that included inpatients with borderline personality disorder (BPD), inpatients with major depression, and a nonpatient control group. When subjects who reported abuse were compared with those who did not, scores for the Global Severity Index (GSI) and all subscales of the SCL-90-R, except for the obsessive-compulsive and somatization subscales, were significantly-higher. When only those subjects who reported CSA were studied and when specific measures of CSA were the independent variables and SCL-90-R subscales were the dependent variables, scores on the hostility, interpersonal sensitivity, and paranoia subscales of the SCL-90-R were significantly higher. Because a large proportion of the sample consisted of borderline patients, and because both the specific measures of CSA and the borderline diagnosis could predict similar SCL-90-R subscale results, a series of stepwise regressions were performed. In the first regression, diagnosis, gender, and specific measures of CSA were the predictor variables and SCL-90-R subscale scores were the dependent variables; in the second regression, SCL-90-R subscales and specific measures of CSA were the predictor variables and diagnosis was the dependent variable. Interpersonal sensitivity was the only significant predictor of the borderline diagnosis. We suggest that, at least in some cases, interpersonal sensitivity may be the constitutional/environmental substrate on which traumatic experiences interact to arrive at the borderline diagnosis.

Prolactin response to morphine in depression

Twenty-two unmedicated inpatients with major depression and 18 healthy volunteers of either sex were given an intravenous injection of 5 mg morphine. Blood samples were drawn immediately before and at intervals for 3 hrs after the injection and assayed for prolactin. Morphine stimulated prolactin secretion. The prolactin response of females was significantly greater than the response of male subjects. There were no significant differences in the prolactin response to morphine between depressed and healthy subjects. The implications of these findings for the hypothesized role of the opioid system in the pathophysiology of depression are discussed.

Prolactin response to TRH in depression

We studied the prolactin response to TRH in 53 unmedicated psychiatric inpatients. The prolactin response of females was significantly greater than the response of male subjects. There was no significant difference in the prolactin response to TRH between depressed patients and those with other psychiatric diagnoses. There was no significant relationship between the prolactin response to TRH and the severity of depression, the TSH response to TRH or the resistance to suppression of cortisol secretion by dexamethasone.

DSM-III and DSM-III-R Schizotypal Symptoms in Borderline Personality Disorder

The frequency of DSM-III and DSM-III-R schizotypal personality disorder (SPD) symptoms and diagnosis was explored in 39 inpatients classified as borderline by the Diagnostic Interview for Borderlines (DIB) and 19 inpatient major depressive disorder (MDD) controls. Most SPD symptoms in all groups, except the nondepressed borderlines, derived from social-interpersonal items. By DSM-III, 24 borderlines (62%) but only six controls (32%) had cognitive-perceptual SPD symptoms (P = .03), whereas by DSM-III-R only 14 borderlines (36%) and seven controls (37%) had such symptoms. Of the 24 borderlines showing cognitive-perceptual symptoms, 16 also had MDD, a significant difference from the non-MDD borderlines (P = .04). This difference disappears in DSM-III-R. The results suggest that some SPD symptoms in borderlines may be related to a concurrent affective episode.

Comparison of early and delayed inpatient dexamethasone suppression tests

Ninety-five inpatients completed a dexamethasone suppression test (DST) within 72 hours after admission and again after at least 1 week of medication-free hospital care. The frequency of cortisol nonsuppression in patients with endogenous depression (ED) was high and not significantly different at both tests. In patients with diagnoses other than ED, the higher rate of cortisol nonsuppression at the first DST was associated with a significant decrease in test specificity. Change in postdexamethasone cortisol levels at repeat testing was associated with a decrease in depressive symptomatology, but was not related to weight change during hospitalization.

Cortisol escape from morphine suppression

Twenty-one unmedicated, sequentially admitted psychiatric patients of either sex and four male healthy volunteers were given an intravenous injection of 2.5 mg morphine. Blood samples were drawn immediately before and at 30-minute intervals for 3 hours after the injection and assayed for cortisol. Morphine suppressed cortisol secretion. Early resumption of cortisol secretion (escape) was more frequent in patients with a diagnosis of major depressive disorder and with abnormal dexamethasone suppression test results. The sensitivity of this infusion paradigm for the diagnosis of major depressive disorder was 40%, and the specificity was 82%. The implications of these findings for the pathophysiology of depression are discussed.