The Association between Attentional and ... - Semantic Scholar

Special Topic Section: Borderline Personality Disorder Editors: J.F. Clarkin, M.I. Posner

Original Paper Psychopathology 2005;38:75–81 DOI: 10.1159/000084814

Received: February 13, 2002 Accepted after revision: October 10, 2003 Published online: March 31, 2005

The Association between Attentional and Executive Controls in the Expression of Borderline Personality Disorder Features: A Preliminary Study Eric A. Fertuck a Mark F. Lenzenweger b John F. Clarkin c a New

York State Psychiatric Institute, Columbia University College of Physicians and Surgeons, New York, N.Y., University of New York at Binghamton, Binghamton, N.Y., and c Weill Cornell Medical College, New York, N.Y., USA b State

Key Words Neurocognition W Attention W Executive function W Borderline personality disorder

Abstract Background: Basic neurocognitive functions such as attention and executive cognitive control represent promising endophenotypes that may improve understanding of the development and expression of borderline personality disorders (BPD). We evaluated the association between performance on the Attention Network Task (ANT) and the Wisconsin Card Sorting Test (WCST), two neurocognitive laboratory instruments, and the extent of BPD psychopathology. Sampling and Methods: We studied 22 BPD-diagnosed individuals who were independently administered these two laboratory assays. Performance on these tests was used as a predictor of the extent of BPD psychopathology. Results: Indexes of the ANT and the WCST were correlated with one another in this sample. Further, the extent of impairment in attention networks, specifically the orienting network, was associated with a greater spectrum of BPD psychopathology, independent of the effects of age and medication status. Finally, ANT and WCST performance were

ABC

© 2005 S. Karger AG, Basel 0254–4962/05/0382–0075$22.00/0

Fax + 41 61 306 12 34 E-Mail [email protected] www.karger.com

Accessible online at: www.karger.com/psp

uncorrelated with general psychosocial functioning in this sample, implicating relative specificity to the extent, as contrasted with functional severity, of BPD psychopathology. Conclusions: Attentional and executive functions are promising endophenotypic markers of BPD psychopathology. The implications of these findings are considered from developmental, experimental, and clinical perspectives. Copyright © 2005 S. Karger AG, Basel

Introduction

Traditionally, borderline personality disorder (BPD) research has relied on clinician interview and self-report questionnaires to assess the manifest phenotypic expressions of borderline psychopathology. More recently, this area of investigation has incorporated the methods of experimental psychopathology [1, 2] and the powerful approaches of the experimental psychology laboratory. Moreover, BPD research approaches have embraced the concept of the endophenotype [3, 4] which has proven useful in the study of schizophrenia. Laboratory-based assessment has allowed for the investigation of endophenotypes in BPD and other psychiatric disorders which are

Eric A. Fertuck, PhD New York State Psychiatric Institute Department of Neuroscience, Unit 42 1051 Riverside Drive, New York, NY 10032 (USA) Tel. +1 212 543 6926, Fax +1 212 543 6946, E-Mail [email protected]

more closely associated with neurophysiology, yet not as observable without focused laboratory assays. Basic neurocognitive functions represent a promising endophenotype that could help to form an improved understanding of development and expression of borderline personality features. Early studies that attempted to relate neurocognitive performance to the BPD diagnosis were hampered by methodological limitations, unfocused instrumentation, and a lack of attention to model-driven hypothesis testing of BPD. More recently, as methodology, measurement, and theory have become more sophisticated, compelling results are emerging. Posner et al. [5] have presented findings that BPD patients, as compared with healthy control participants, are specifically deficient in a neural network associated with the regulation of executive control. These researchers utilized the Attention Network Task (ANT) which assesses three independent attentional functions. The first is alerting, the capacity to sustain an alert cognitive state. The second is orienting which involves focused identification and selection of sensory stimuli. The third is conflict (executive control) which is the capacity to decide among competing responses based upon a principle or goal. The BPD participants were specifically impaired relative to control participants in the executive control network, as assessed by the ANT, but did not differ from the control group on the alerting and orienting tasks. Further, a group of individuals who exhibited temperamental features similar to BPD, but did not meet criteria for BPD, performed better than the BPD group on the conflict index of the ANT. However, the temperamentally matched group was not significantly different from the other two groups on the conflict index. The anterior cingulate is a brain region which is essential to executive controls as measured by a conflict flanker task [6]. Consistent with these findings, BPD patients have also demonstrated deficits in executive functions on the Wisconsin Card Sorting Test (WCST) [7] as compared with healthy control subjects. These same BPD participants showed no significant differences as compared with control participants on other measures of working memory and sustained attention [8]. The WCST is a well-established measure of executive functions that evaluates an individual’s capacity to adhere to a cognitive principle and respond in a manner consistent with this principle. Further, it assesses the individual’s capacity to ascertain shifts in rules during the task and respond to the new rule in a consistent manner. WCST performance has most frequently been implicated in frontal lobe function and dysfunction associated with the dorsolateral prefrontal cor-

76

Psychopathology 2005;38:75–81

tex; however, many studies have pointed to a more systemic brain activation [9]. The tasks of the WCST are, then, consistent with executive control, but in a manner that may complement conflict scores of the ANT. Thus, two studies implicate specific deficits in executive control using different instruments in the expression of BPD. Given the broad set of functions associated with effortful control, we sought to correlate the ANT with performance on the WCST among a sample of 22 BPD subjects who completed both the ANT and the WCST. Additionally, we sought to evaluate the association between performance on ANT and WCST and the number of diagnostic criteria for BPD met in these same 22 BPD patients. The extent of BPD pathology is not necessarily synonymous with the level of psychosocial functioning which is often assessed using the Global Assessment of Functioning (GAF) Scale of the DSM-IV [10]. To investigate the specificity of attentional and executive functions in the extent of BPD pathology, we will also correlate neurocognitive performance with GAF scores in this BPD sample. This study builds upon previous research by bringing together data on the ANT and the WCST in a study of BPD-diagnosed subjects who met DSM-IV criteria for the disorder. This within-group study contrasts with the between-group-designed studies reviewed above. In this study, we addressed the following questions. (1) Are ANT conflict scores, a measure of effortful control, associated with deficits on neighboring tests/functions of effortful control, such as the WCST, in a BPD sample? We hypothesized that the conflict dimension of the ANT would be associated with the WCST dimensions, as both differentiate BPD from healthy control individuals. (2) Is the degree of impairment on neurocognitive functions associated with the extent of BPD symptomatology? We hypothesized that deficits in the ANT conflict and WCST dimensions would be related to an increased number of BPD criteria, but not with global psychosocial functioning (GAF). We also explored whether the alerting and orienting dimensions of the ANT were related to the degree of BPD criteria.

Patients and Methods Patients All patients completed a comprehensive series of assessment interviews, including an evaluation of axis I diagnoses. Twenty-two female BPD patients were recruited through the Weill Cornell BPD treatment study prior to the beginning of their treatment. Table 1 summarizes ethnic/racial distribution, educational level, and treat-

Fertuck/Lenzenweger/Clarkin

ment history of the sample. Inclusion criteria included meeting DSM-IV [10] BPD criteria on the International Personality Disorders Examination (IPDE) [11] and being between the ages of 18 and 50 years. The mean age of this sample was 32.3 B (SD) 9.2 years. Reliability coefficients for the number of BPD DSM-IV criteria met (estimate for a single rater) were ICC(1,1) = 0.825 based on a comparison of two raters who independently diagnosed an n of 43 from the BPD treatment sample. The treatment sample is the same sample from which the sample of 22 for this study was derived. Exclusion criteria included any psychotic disorder, substance dependence, mental retardation, and/or bipolar I disorder. Additionally, we excluded any participant with a major depressive episode, eating disorder, and/or an episode of substance dependence during the 6 months prior to assessment. All axis I and substance abuse disorders were assessed with the SCID-I [12] by trained raters, and reliability scores for unipolar affective diagnosis were Kappa (SD) = 0.739 (0.122), for bipolar affective diagnosis Kappa (SD) = 0.783 (0.208), for alcohol and/or substance dependence diagnosis Kappa (SD) = 1.000 (0.000), for anxiety disorder diagnosis Kappa (SD) = 0.587 (0.137), and for eating disorder diagnosis Kappa (SD) = 0.901 (0.097). With respect to these diagnoses, 86.4% of the sample met the criteria for at least one current axis I condition. Specific current axis I disorder frequencies in the sample were as follows: dysthymia 32%, depressive disorder (not otherwise specified) 4.5%, panic disorder without agoraphobia 4.5%, panic disorder with agoraphobia 4.5%, social phobia 13.6%, obsessive-compulsive disorder 9.1%, posttraumatic stress disorder 4.5%, generalized anxiety disorder 18.2%, and eating disorder (not otherwise specified) 36.4%. This general pattern of comorbid axis I condition is not uncommon for BPD patients. A history of past axis I disorders is consistent with this general pattern for current disorders, but these data were omitted to conserve space. Measures Wisconsin Card Sorting Test. The WCST [7] is a well-known neuropsychological measure of executive cognitive function which is associated with dorsolateral prefrontal cortex functioning. A computerized version of the WCST was used. Attention Network Task. The ANT assesses the efficiency of three attentional control systems [13]. The ANT measures efficiency by assessing cued reaction time and performance during a flanker task. The ANT is a computer-administered test. BPD Dimensional Score. The IPDE [11] generates three scores for each personality disorder. First, there is a categorical score indicating whether a subject meets criteria for the disorder. All the subjects in this sample scored positive for BPD on the IPDE by DSM-IV criteria. Second, the IPDE calculates the number of criteria for the disorder met, which for BPD is between 5 and 9. Third, the IPDE calculates a dimensional score. The dimensional score is based upon a sum of the 0, 1, or 2 score for each BPD criterion, with 0 representing absence of the criterion, 1 being subthreshold, and 2 indicating presence of the criterion. Since any combination of 5 or more of the 9 criteria would qualify an individual for the BPD diagnosis, a BPD dimensional score between 10 to 18 is possible. This number is an index of extent of BPD features as indicated by the number of criteria met and has been extensively used in previous published studies on BPD and other personality disorders [14–16]. The reliability for the IPDE-BPD dimensional rating (estimate for a single rater) was ICC(1,1) = 0.863 (n = 43), as assessed in a larger group of BPD participants from which the current sample was derived.

Attention, Executive Control, and Borderline Personality

Table 1. Demographic and treatment history characteristics for

BPD patients (n = 22) Characteristics

n

%

Ethnic/Racial Distribution Caucasian African-American/black Asian Mixed Other

13 5 1 2 1

59.1 22.7 4.5 9.1 4.5

Educational level High school or equivalent Some college Bachelor’s degree Graduate or postgraduate

2 5 9 6

9.1 22.7 40.9 27.3

1 11 6

4.5 50.0 27.3

Treatment history Past detox/substance abuse treatment – lifetime Psychiatrically hospitalized – lifetime Psychiatrically hospitalized – past year

Procedures To reduce the amount of stress and anxiety involved in participating in the interview and assessments, the administration of the diagnostic interviews, the WCST, and the ANT were on different days, respectively. The participants were paid for their participation.

Results

Table 2 summarizes the mean values and standard deviations of the ANT and WCST scores among these 22 BPD participants. The ANT scores were correlated with WCST dimensions (table 3). Consistent with hypotheses, more impaired, higher ANT alertness scores were correlated with more percent perseverative errors and responses on the WCST. By contrast, more impaired, higher ANT conflict scores were correlated with increased percent nonperseverative errors and fewer conceptual level responses on the WCST. A partial correlation or the ANT and WCST variables using raw WCST scores (not age and education corrected) controlling for age demonstrated that age influences some of the correlations, but does not eliminate them entirely. In our next analyses, we investigated the associations between the dimensions of the ANT and WCST and the degree of BPD features, as assessed by the BPD dimensional score of the IPDE. Partially consistent with hypoth-


77

Table 2. Descriptive statistics of ANT, WCST, and BPD dimension-

Table 5. Exploratory forced-entry regression analysis of ANT di-

al score variables for BPD patients (n = 22)

mensions in relation to the BPD dimensional score (n = 22) controlling for age

Test variable

Mean

ANT alerting (adjusted) ANT orienting (adjusted) ANT conflict (adjusted) WCST standard score for nonperseverative errors WCST standard score for percent nonperseverative errors WCST standard score for percent conceptual level responses WCST standard score for percent perseverative errors WCST standard score for percent perseverative responses WCST standard score for perseverative errors BPD dimensional score

Standard deviation

0.07 0.11 0.26

0.05 0.05 0.11

96.90

11.29

95.55

13.30

94.23

12.73

92.91

13.48

93.18

14.00

94.27 14.73

12.42 2.53

Table 3. Correlations between ANT and WCST and BPD dimensional scores (n = 22)

WCST variables (standard scores)

ANT variables alerting

orienting

conflict

% perseverative responses % perseverative errors % nonperseverative errors % conceptual level responses

0.43* 0.39* 0.24 0.30

–0.30 –0.31 –0.11 –0.29

0.24 0.27 0.38* 0.38*

BPD dimensional score

0.40*

0.27

ANT scores adjusted for reaction time; WCST standard scores are age and education corrected. *p ! 0.05 (one-tailed).

Table 4. Exploratory forced-entry regression analysis of ANT di-

mensions in relation to the BPD dimensional score (n = 22) ANT

B

SEB

ß

pr

t

p

Conflict Orienting Alerting

8.99 25.37 18.42

4.36 9.31 9.36

0.38 0.50 0.35

0.27 0.52 0.42

2.06 2.73 1.97

0.05 0.01 0.07

R² = 0.45; pr = part (or semipartial) correlation coefficient.

78


B

SEB

ß

pr

t

p

Conflict Orienting Alerting

7.41 26.26 16.05

4.37 9.07 9.24

0.31 0.52 0.30

0.19 0.56 0.39

1.70 2.90 1.74

0.11 0.01 0.10

R² = 0.51; pr = part (or semipartial) correlation coefficient.

WCST standard scores are age and education corrected.

0.41*

ANT

eses, poorer performances on both the alertness and orienting scores, but not the conflict score, were significantly correlated with higher levels of BPD symptomatology (table 3). The WCST variables, in contrast to the ANT, were not related to the extent of BPD symptomatology, counter to our hypothesis. In a linear regression analysis, with BPD dimensional score as the dependent variable, both poorer performances on conflict and orienting scores contributed unique variance in the regression model to predict higher levels of BPD symptomatology (table 4). We also conducted this regression model controlling for the effects of age. In this regression, the conflict score was no longer predictive of the BPD dimensional score. However, the orienting score still predicted the BPD dimensional score at the p ! 0.01 level (table 5). This suggests that age is a factor that accounts for some of the variance in the association between the conflict in relation to the BPD dimensional score. In this context, it is important to point out that within this BPD group, a subgroup of BPD subjects (n = 15) was medicated, whereas the remainder (n = 7) were not. A comparison of the performance of the medicated versus the nonmedicated BPD subjects did not reveal a single statistically significant differences on the ANT and WCST performance indexes (p ! 0.05, two-tailed), with one exception. The conflict index of the ANT was significantly higher in the medicated (mean = 0.30; SD = 0.11) as compared with the unmedicated group [mean = 0.18; SD = 0.05; t(20) = –2.82, p ! 0.01, two-tailed]. However, a regression model controlling for medication status of the BPD participants did not significantly alter the original model, suggesting that the ANT findings cannot be accounted for by whether or not a BPD individual was taking medication.


In order to evaluate the possibility that our findings simply indicate a correlation between neurocognitive performance and global psychosocial functioning, we correlated the GAF score (mean = 47.3; SD = 8.9) of the SCIDI in this sample with the ANT and WCST variables and found that there were no significant correlations between these neurocognitive indices and GAF scores at the p ! 0.05 level (two-tailed). Further, the BPD dimensional score was not correlated with the GAF score. It appears that the performance on the ANT is more specifically related to the extent of BPD features and not to the global level of psychosocial functioning in this BPD sample.

Discussion

This study of BPD individuals demonstrates that indexes within two separate laboratory assays of executive and attentional control, respectively, were correlated with one another. Specifically, we found that poorer performance on the conflict score of the ANT was significantly correlated with increased nonperseverative errors and fewer conceptual-level responses on the WCST. This finding reinforces the association of BPD with executive control impairments, as these indexes on the ANT and WCST both differentiated BPD individuals from control groups in previous studies. We also found that a poorer performance on the alerting dimension of the ANT was associated with both greater percent perseverative errors and responses on the WCST. Though unexpected, this finding is theoretically interpretable, since perseverative errors can indicate a failure to maintain a rule during the WCST. This failure can be caused by poor sustained attention over the course of the WCST task. We also found that the extent of impairment in attention networks was associated with a greater spectrum of BPD pathology, as indicated by the BPD dimensional score, but not with the level of psychosocial functioning as assessed by the GAF. Poorer performances on the conflict and orienting tasks of the ANT were associated with more BPD symptoms, as assessed by independent semistructured clinical interview. The impairment in the orienting network was associated with a greater spectrum of BPD features independent of the effect of age and medication status, but the conflict score was no longer predictive of the dimensional score after controlling for age. It could be that a restricted range in the conflict scores in this sample contributed to the lack of significance in between the conflict score of the ANT and the WCST performance and the BPD dimensional score.


This study also provides initial support for the construct validity of the ANT task through the significant association with the more psychometrically established WCST indexes. Conversely, the associations between the alerting network on the ANT and perseverative errors on the WCST reinforce the notion that the WCST assesses a broad range of function, beyond only executive control. It was unexpected that the WCST executive control indexes that distinguish BPD from control participants, such as percent perseverative and nonperseverative errors, do not appear to be associated with the extent of BPD pathology. This contrasts with the conflict attention network in the ANT which was associated with the extent of BPD pathology and also significantly distinguishing BPD from control participants [5]. The differences between these tasks may point to one potential explanation. The three ANT indexes were designed to assess functionally different networks at both behavioral and neurophysiological levels. By contrast, the WCST involves the coordinated, voluntary control of many neurocognitive functions such as working memory, attention, and inhibitory controls. Since the WCST performance is not consistently associated with any distinctive constellation of neurophysiological activation, we may speculate that with the more focused differentiation of tasks in the ANT, compelling associations between the orienting network, and to a smaller extent the conflict network, and BPD features could emerge. Insofar as these ANT tasks are attentional and are not directly implicated in emotional systems – the dysregulation of which is putatively central to the expression of BPD features – this is a striking association. Further investigations into the interrelationships between attentional, executive, and emotion systems in BPD would seem warranted. These laboratory-based findings are similar to those of Hoermann et al. [17]. These authors, using self-report measures of executive control, found that different clusters of BPD patients could be differentiated in terms of severity of symptoms and other personality variables based upon effortful control self-report ratings on the Adult Temperament Questionnaire. Our laboratorybased findings reinforce the results of this study. However, the study performed by Hoermann et al. [17] utilized only a measure of effortful (executive) control, and our study also found that the orienting network was associated with BPD features, suggesting that laboratorybased tasks may have the unique capacity to tap salient BPD endophenotypes as compared with self-report ratings.


79

A limitation of the BPD dimensional score used in this study is that it does not differentiate subtypes of BPD features, as they factor together in factor-analytic studies [18]. In these studies BPD features consistently factor themselves into three groups: affective instability, selfdisturbance and problematic interpersonal relating, and impulsive-aggression. Because the BPD dimensional scores do not capture these subdimensions, we could not associate the WCST and ANT with specific subtype features. For example, previous studies suggest that impaired executive functions are perhaps uniquely associated with impulsive suicidal and parasuicidal behaviors [19]. The investigation of the transactional relationship between development, endophenotype, and phenotype represents a major challenge in BPD research. Given the preliminary nature of this study, its small sample size, and the all female nature of the sample, more questions are raised than answered regarding how attention, executive control, and emotional dysregulation mutually influence one another in the development and expression of BPD. For example, what is the relationship between the attentional networks and the heterogeneous mixture of BPD features such as identity disturbance, dissociative phenomena, impulsivity, fear of abandonment, and affective instability? We may speculate that this attentional networks should be more associated with cognitive-perceptual disturbances such as dissociation and quasi-psychotic thought, since basic attentional weaknesses underly such symptoms in other disorders such as schizotypy [20]. Theory-driven studies addressing such questions could begin to isolate aspects of BPD phenomenology to address this and similar research questions. Another important area of investigation would be how basic neurocognitive functions interrelate with the attachment system and other higher-order personality variables such as identity and moral values. Executive and attentional systems, theoretically, must operate in concert with autobiographical memory, affective-motivational systems, and interpersonal, developmental experience to shape enduring psychological structures that embody the sense of self in relation to its environment. The interface between neurocognition and social cognition would appear particularly relevant. BPD individuals appear to exhibit a wide range of social cognitive difficulties such as inaccurately inferring the meaning of social cues, experiencing interpersonal relationships as need gratifying and exploitative, quickly idealizing and depreciating important people in their lives, and exhibiting significant variability in moral values, vocational commitment, and in their own sense of what makes them and others unique as

80


individuals [21]. One could hypothesize that neurocognitive vulnerabilities contribute to social cognitive vulnerabilities, yet the two have not been studied concurrently. Finally, a developmental perspective can shed light upon the precursors and risk factors, genetic and psychosocial, of BPD. The few studies in this area that have been conducted suggest that neurocognitive weakness (on the WCST) and adverse psychosocial experience interact to predict the development of BPD features in children [22]. Preliminary longitudinal studies of BPD have suggested that up to two thirds of the young adult BPD individuals recover or exhibit diminished symptoms, starting at approximately age 30 years [23]. Moreover, in studies of healthy individuals, executive cognitive controls diminish with age, beginning in middle adulthood [24]. The relationship between cognitive functions, age, and BPD features remains a novel, yet potentially fruitful area in BPD basic research of developmentally influenced psychobiological markers of the disorder. The decline of BPD features with age may explain why controlling for age in our sample influenced the prediction of poorer conflict network performance and the extent of BPD features. The ongoing study of neurocognitive functions in BPD and its life course could inform us regarding the developmental psychopathology of the disorder at both ends of the life span. The clinical implications of research into these functions in BPD warrant theory-driven investigation as well. Neurocognitive vulnerabilities likely influence the response to psychosocial and pharmacological interventions, in particular when they occur in the context of a personality disorder. Since information processing is often compromised in BPD, learning and behavior change may take longer. Indeed, comorbidity of an axis II disorder is a negative predictor of response to treatment of axis I conditions [25]. It seems quite probable that differing patterns of neurocognitive and social cognitive strengths and weaknesses will respond differentially to different treatment approaches. For example, exploratory psychotherapy likely requires a minimal level of attention and executive control to be able to utilize the here-and-now understanding of intense emotional states with the therapist. Additionally, from a clinical research standpoint, with well-established endophenotypic markers of BPD, we may eventually have at our disposal improved measures of treatment outcome that could be more sensitive and valid measures of enduring change than manifest symptom and personality measures that are now commonly utilized. Finally, with well-established markers and understanding of differential treatment responses, the


diagnostic criteria for BPD can be meaningfully revisited. This would allow for sophisticated differential treatment planning for BPD and its potential subtypes, with the goal of minimizing treatment failures and impasses and maximizing both symptom reduction and quality of life for individuals struggling with this devastating syndrome.

Acknowledgments This research was supported by the Borderline Personality Disorder Research Foundation and the Leslie Glass Foundation. We would like to thank Michael I. Posner and his colleagues for providing us with the ANT data. Further, Michael I. Posner provided valuable feedback on earlier versions of this article. We would also like to acknowledge the data collection and management assistance of Kenneth Critchfield, Jill Delaney, Simone Hoermann, Kenneth N. Levy, Joel McClough, and Natalie Vizueta.

References 1 Lenzenweger MF, Hooley JM: Principles of experimental psychopathology: Essays in honor of Brendan A. Maher. Washington, APA, 2003. 2 Korfine L, Hooley JM: Directed forgetting of emotional stimuli in borderline personality disorder. J Abnorm Psychol 2000;109:214–221. 3 Gottesman II, Gould TD: The endophenotype concept in psychiatry: Etymology and strategic intentions. Am J Psychiatry 2003;160:636– 645. 4 Gottesman II, Shields J: Schizophrenia and Genetics: A Twin Study Vantage Point. New York, Academic Press, 1972. 5 Posner MI, Rothbart MK, Vizueta N, Levy K, Evans DE, Thomas KM, Clarkin J: Mechanisms of borderline personality disorder. Proc Natl Acad Sci U S A 2002;99:16366–16370. 6 Botwinick MM, Braver TS, Barch DM, Carter CS, Cohen JD: Conflict monitoring and cognitive control. Psychol Rev 2001;108:624–652. 7 Heaton RK: Wisconsin Card Sorting Test Manual. Odessa, Psychological Assessment Resources, 1981. 8 Lenzenweger MF, Clarkin JF, Fertuck EA, Kernberg OF: Executive neurocognitive functioning and neurobehavioral systems indicators in borderline personality disorder: a preliminary study. J Personal Disord 2004;18: 421–438.


9 Strauss E, Hunter M, Wada J: Wisconsin Card Sort performance: Effects of age of onset and laterality of dysfunction. J Clin Exp Neuropsychol 1993;15:896–902. 10 American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, ed 4. Washington, American Psychiatric Association, 1994. 11 Loranger AW: International Personality Disorder Examination (IPDE) Manual. Odessa, Psychological Assessment Resources, 1999. 12 First MB, Spitzer RL, Gibbon M, Williams JB: User’s Guide for the Structured Clinical Interview for DSM-IV Axis I Disorders – Clinical Version (SCID-CV). Washington, American Psychiatric Press, 1997. 13 Fan J, McCandliss BD, Sommer T, Raz M, Posner MI: Testing the efficiency and independence of attentional networks. J Cogn Neurosci 2002;3:340–347. 14 Lenzenweger MF: Stability and change in personality disorder features: The Longitudinal Study of Personality Disorders. Arch Gen Psychiatry 1999;56:1009–1015. 15 Loranger AW, Lenzenweger MF, Gartner AF, Susman VL, Herzig J, Zammit GK, Gartner JD, Abrams RC, Young RC: Trait-state artifacts and the diagnosis of personality disorders. Arch Gen Psychiatry 1991;48:720–728. 16 Loranger AW, Sartorius N, Andreoli A, et al: The International Personality Disorder Examination: The World Health Organization/Alcohol, Drug Abuse, and Mental Health Administration international pilot study of personality disorders. Arch Gen Psychiatry 1994;51:215– 224.

17 Hoermann S, Clarkin JF, Hull J, Levy KN: The construct of effortful control: An approach to borderline personality disorder heterogeneity. Psychopathology 2005;38:82–86. 18 Clarkin JF, Widiger T, Frances A, Hurt SW, Gilmore M: Prototypic typology and the borderline personality disorder. J Abnorm Psychol 1983;92:263–275. 19 Keilp JG, Sackeim HA, Brodsky BS, Oquendo MA, Malone KM, Mann JJ: Neuropsychological dysfunction in depressed suicide attempters. Am J Psychiatry 2001;158:735–741. 20 Lenzenweger MF: Schizophrenia: Refining the phenotype, resolving endophenotypes. Behav Res Ther 1999;37:281–295. 21 Westen D: Social cognition and object relations. Psychol Bull 1991;109:429–455. 22 Zelkowitz P, Paris J, Guzder J, Feldman R: Diatheses and stressors in borderline pathology of childhood: The role of neuropsychological risk and trauma. J Am Acad Child Adolesc Psychiatry 2001;40:100–105. 23 Stone MH: Long-term outcome in personality disorders. Br J Psychiatry 1993;162:299–313. 24 West RL: An application of prefrontal cortex function theory to cognitive aging. Psychol Bull 1996;120:272–292. 25 Clarkin JF: Treatment of personality disorders. Br J Clin Psychol 1996;35:641–642.


81