Interpersonal functioning in borderline personality disorder: A systematic review of behavioral and laboratory-based assessments

Clinical Psychology Review 34 (2014) 193–205

Contents lists available at ScienceDirect

Clinical Psychology Review

Interpersonal functioning in borderline personality disorder: A systematic review of behavioral and laboratory-based assessments Sophie A. Lazarus a,⁎, Jennifer S. Cheavens a, Francesca Festa b, M. Zachary Rosenthal c a b c

Department of Psychology, The Ohio State University, USA School of Education, The Ohio State University, USA Department of Psychiatry and Behavioral Sciences, Duke University Medical Center, USA

H I G H L I G H T S • • • • •

Interpersonal dysfunction is central in borderline personality disorder (BPD). We highlight objective measures of areas of potential impairment in those with BPD. Those with BPD have heightened emotional reactivity to interpersonal stressors. Individuals with BPD show impairment in trust and cooperation. We offer suggestions for future research.

a r t i c l e

i n f o

Article history: Received 6 May 2013 Revised 18 January 2014 Accepted 23 January 2014 Available online 5 February 2014 Keywords: Borderline personality disorder Interpersonal functioning Social cognition Trust/cooperation Parenting Interpersonal aggression

a b s t r a c t It is widely accepted that interpersonal problems are a central area of difficulty for those with borderline personality disorder (BPD). However, empirical elucidation of the specific behaviors, or patterns of behaviors, characterizing interpersonal dysfunction or dissatisfaction with relationships in BPD is limited. In this paper, we review the literature on interpersonal functioning of individuals with BPD by focusing on studies that include some assessment of interpersonal functioning that is not solely self-report; that is, studies with either behavioral laboratory tasks or manipulation of interpersonal stimuli in a controlled laboratory setting were included. First, we review the literature relevant to social cognition, including perceptual biases, Theory of Mind/empathy, and social problem-solving. Second, we discuss research that assesses reactivity to interpersonal stressors and interpersonal aggression in BPD. Next, we review the literature on trust and cooperation among individuals with BPD and controls. Last, we discuss the behavior of mothers with BPD in interactions with their infants. In conclusion, we specify areas of difficulty that are consistently identified as characterizing the interpersonal behaviors of those with BPD and the relevant implications. We also discuss the difficulties in synthesizing this body of literature and suggest areas for future research. © 2014 Elsevier Ltd. All rights reserved.

Contents 1. 2.

3.

Introduction . . . . . . . . . . . . . . . . . . . Characterization of interpersonal dysfunction in BPD 2.1. Social cognition . . . . . . . . . . . . . . 2.1.1. Perceptual biases . . . . . . . . . 2.1.2. Theory of Mind (TOM) and empathy 2.1.3. Social problem-solving . . . . . . 2.2. Reactivity to interpersonal stressors . . . . . 2.3. Interpersonal aggression . . . . . . . . . . 2.4. Lack of cooperation/trust . . . . . . . . . . 2.5. Behavior within mother–child interactions . Discussion . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

⁎ Corresponding author at: 181 Psychology Building, 1835 Neil Avenue, Columbus, OH 43210, USA. E-mail address: [email protected] (S.A. Lazarus). http://dx.doi.org/10.1016/j.cpr.2014.01.007 0272-7358/© 2014 Elsevier Ltd. All rights reserved.

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

194 195 195 196 197 198 198 199 200 200 201

194

S.A. Lazarus et al. / Clinical Psychology Review 34 (2014) 193–205

3.1. Future directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Appendix A. Supplementary data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction Borderline personality disorder (BPD) is a serious public health problem that poses considerable challenges for mental health professionals, those suffering from the disorder, and their families. Individuals with BPD disproportionately present for treatment in both inpatient and outpatient clinics relative to many other disorders, leading to high rates of health care utilization and associated costs (Skodol et al., 2005). Although the diagnostic criteria for BPD (American Psychiatric Association, 2013) include dysfunction across a wide range of neurobehavioral systems, including emotional expression (e.g., marked reactivity), behavioral inhibition (e.g., impulsivity), cognition (e.g., paranoia or dissociation when acutely distressed), and interpersonal functioning (e.g., fear of abandonment), disturbed interpersonal relationships are increasingly being recognized as central to understanding the impairments and psychological distress associated with the disorder (Gunderson, 2007). The empirical investigation of interpersonal functioning in BPD occurs in the context of a rich theoretical history. Several interrelated psychodynamic and psychoanalytic theories explain interpersonal disturbances in BPD. These theories emerged as clinicians observed the centrality of interpersonal problems for individuals with BPD and endeavored to explain the origin of pathological and extreme interpersonal behavior, such as suicidal behaviors in response to interpersonal conflict or rejection. While these theories generally focus on problems in early relationships and caregiving experiences, each has a somewhat unique explanation for the development of dysfunctional interpersonal behaviors in adulthood. In one such interpersonal theory of BPD, object relations theorists (e.g., Jacobson, 1964; Kernberg, 1980; Klein, 1957) posit that selfother representations form in early relationships, particularly between the infant/child and the primary caregiver, and that these cognitive representations play a central role in personality development. Some have argued (e.g., Westen, 1991) that the emotions and expectations attached to these representations are critically important determinants of functioning in interpersonal relationships as dyads are linked by the affective valence of the representations. For example, according to Clarkin, Lenzenweger, Yeomans, Levy, and Kernberg (2007), individuals with BPD have representations of self and others that are affectively split (i.e., positive and negative representations) and lack integration (i.e., unstable representations). Thus, object relations theorists would predict polarized interpretations of others (e.g., dichotomous thinking), which are heavily influenced by the affect linking the dyad within the particular interaction. Another way of understanding the interpersonal behavior associated with BPD is through the lens of attachment theory. Attachment theorists posit that children, based mostly on interactions with primary caregivers, develop internal models of the self and others that guide expectations and beliefs in relationships, particularly in times of stress (Bowlby, 1973). Secure attachment with the caregiver allows the child to develop and maintain a coherent and positive sense of self and expectations for responsive and caring behavior from others. In contrast, BPD is typically characterized by disturbed attachment and representations of the self and others that are inconsistent and negative (Agrawal, Gunderson, Holmes, & Lyons-Ruth, 2004). According to Fonagy, Target, Gergley, Allen, and Bateman (2003), the development of secure attachment hinges on caregivers' abilities to understand their own and others' minds and help the child develop this capacity (i.e., provide a scaffolding for mentalization). A failure to develop the ability to perceive and interpret behavior based on underlying mental states

202 203 203

(mentalization) may lead to difficulty interpreting and understanding interpersonal experiences, especially in contexts where the attachment system is activated (i.e., under conditions of perceived threat). Accordingly, this theory predicts that deficits in mentalization associated with maladaptive attachment account for the interpersonal dysfunction among individuals with BPD. Linehan's biosocial model (Linehan, 1993) is an alternative account of the development of interpersonal problems in BPD. According to this model, an underlying biological vulnerability to emotional dysregulation (i.e., high sensitivity and reactivity to emotional stimuli, slow return to baseline after emotional arousal) transacts with environmental stressors (i.e., invalidation) to contribute to emotional and interpersonal impairments. The transactional interplay between these biological and social factors is believed to adversely influence the development of one's sense of self and other, disrupting the development of healthy relationships. Thus, in this model, disrupted (or less than ideal) relationships function as both a risk factor for the development of BPD and a consequence of the disorder. The consistent undermining of one's internal experience (i.e., invalidation) may interfere with healthy interpersonal relations by contributing to a disturbed learning history for close relationships, creating an overreliance on others' opinions and indications of worth, and encouraging dichotomous (i.e., all good or all bad) thinking about others. Consistent with interpersonal theories of BPD, evidence that interpersonal functioning is a major area of concern for those with BPD can be found across converging areas of empirical research. For example, factor analytic studies indicate that disturbed interpersonal relations represent a key factor underlying the variance across BPD symptoms (Sanislow et al., 2002). Further, individuals with BPD often report greater problems with interpersonal functioning compared to healthy controls (e.g., Bouchard, Sabourin, Lussier, & Villeneuve, 2009). Additionally, some of the most serious outcomes related to BPD, such as self-injury and suicide, frequently occur in interpersonal contexts (e.g., Brodsky, Groves, Oquendo, Mann, & Stanley, 2006; Brown, Comtois, & Linehan, 2002) and are related to problems with social adjustment (Soloff & Fabio, 2008). Prospective studies suggest that improvement in interpersonal functioning occurs more gradually in BPD than in several other Axis II disorders (Choi-Kain, Zanarini, Frankenburg, Fitzmaurice, & Reich, 2010; Skodol et al., 2005). In fact, certain interpersonal symptoms such as negative affect when alone, fear of abandonment, discomfort with care, and dependency are extremely slow to remit, with 15% to 25% of individuals with BPD who exhibited these symptoms at baseline failing to show improvement at 10-year follow-up (Choi-Kain et al., 2010). Further, remission from the disorder is often related to positive interpersonal events, such as entering a stable relationship (Links & Heslegrave, 2000). Thus, impairment in interpersonal functioning: (a) is theoretically and diagnostically central to BPD, (b) is associated with self-injurious behavior and other adverse clinical outcomes, (c) plays an important role in the prognosis and course of BPD, and (d) is reported by those with BPD as significantly problematic. The evidence clearly suggests that interpersonal functioning in BPD is often meaningfully impaired. What is less clear, based on the existing body of research, is how to precisely characterize the various interpersonal impairments in BPD. In recent years, the pace of empirical research examining problems with interpersonal functioning in BPD has accelerated. The field has moved from a primary reliance on cross-sectional self-report to more sophisticated designs using prospective methodologies and more ecologically valid assessments of interpersonal behavior. The use of such

S.A. Lazarus et al. / Clinical Psychology Review 34 (2014) 193–205

laboratory-based methods of measurement has much to offer the study of interpersonal functioning in BPD. When characterizing interpersonal impairments in those with BPD, it is useful to have multiple methods of assessment for specific behaviors to identify the basic processes that might differentiate them from others. This also affords the opportunity to assess how self-reported interpersonal problems align with laboratory-based measures of these same difficulties. In addition, assessments that more closely approximate real world interpersonal settings can advance an empirical understanding of the interpersonal contexts (e.g., social rejection) that elicit BPD criterion behaviors (e.g., hostility/anger). Another advantage of focusing on behavior in specific contexts is that this approach mitigates some of the challenges associated with retrospective self-reports of general interpersonal functioning (e.g., self-report biases and inaccuracies), which will be discussed below. Put simply, these objective and performance-based measures, when used in conjunction with traditional self-report and interview measures, may advance the precision with which we understand how to characterize interpersonal problems in BPD. There have been few attempts to review and synthesize the existing research base on interpersonal functioning in BPD (for exceptions, see Bornstein, Becker-Matero, Winarick, & Reichman (2010) for a review of dependency in BPD and Agrawal et al. (2004) for a review of attachment in BPD). To our knowledge, no reviews have focused solely on behavioral and laboratory assessments of interpersonal functioning in BPD. In an attempt to maintain a focus on objective measures of the interpersonal functioning of those with BPD, we included only empirical studies that a) sampled individuals with BPD symptoms or diagnoses and b) included objective measures of either interpersonal behavior or responses to interpersonal assessments in controlled laboratory settings. We included findings derived from subjective responses to interpersonal stimuli in laboratory behavioral paradigms; however, studies that included only self-report measures were excluded for two reasons. First, self-report measures tend to assess habitual responding or “average” states and we were interested in specific instances of interpersonal behavior. Second, there is evidence to suggest that people with BPD have disproportionate access to negative memories and are likely to hold negative evaluations of themselves and others (Baer, Peters, Eisenlohr-Moul, Geiger, & Sauer, 2012). As such, relying solely on selfreports of interpersonal functioning may obscure potentially important variations in interpersonal behaviors. In this review, we summarize the research in an attempt to characterize difficulties with interpersonal behavior among individuals with BPD. Our goal is to integrate the existing literature in this area, generate hypotheses for further study, and stimulate future research focused on understanding the interpersonal disturbances among individuals with BPD. Ultimately, as interpersonal difficulties in BPD become better elucidated, the next generation of assessment measures and interventions can be developed to more rapidly and directly target specific interpersonal problems for individuals with this disorder. 2. Characterization of interpersonal dysfunction in BPD The existing literature can be organized into several specific areas of potential impairment. These domains of interpersonal dysfunction, introduced below, provide an atheoretical and empirically-based framework for the scope of this literature review.1 The first, social cognition, includes studies of perceptual biases, Theory of Mind (TOM), and social problem-solving. Research on interpersonal perceptual biases addresses the question of whether individuals with BPD can be characterized by dysfunctional appraisals of others. TOM research examines the ability of those with BPD to recognize and understand interpersonal situations, including the anticipated emotional reactions of others. Social problemsolving research aims to examine whether individuals with BPD 1 See supplementary information for tables detailing the studies included in each section.

195

demonstrate impairments in problem-solving skills or cognitive flexibility when encountering social problems. Thus, tests of social cognition would suggest that one way to characterize interpersonal dysfunction in BPD is via differences in interpretations of interpersonal situations. A second domain of potential interpersonal dysfunction in BPD is reactivity to interpersonal stressors. In these studies, researchers use laboratory biobehavioral methods in an attempt to characterize different patterns of reactivity among individuals with BPD and controls to both interpersonal and non-interpersonal stimuli. Stronger reactions to interpersonal stimuli, compared either to non-interpersonal stimuli or controls, might suggest that heightened negative emotion in interpersonal situations characterizes BPD. The third domain of interpersonal impairment addressed in this review is interpersonal aggression. Interpersonal aggression, particularly in response to rejection cues, might increase the probability of unstable or otherwise problematic relationships for those with BPD. The fourth domain of interpersonal impairment discussed examines research aimed at identifying observable differences in trust and cooperation between those with BPD and others. If BPD is characterized by a lack of trust and cooperation, it is likely to result in an impaired ability to participate in and make full use of social networks and relationships. Last, we consider a relatively recent body of research that uses laboratory-based interactions between mothers and infants to identify how women with BPD may differ from others in behavior towards their children. If women with BPD are indeed characterized by specific patterns of interpersonal behavior with their children, this information could inform models of both the transmission and maintenance of BPD features. 2.1. Social cognition Successfully navigating interpersonal interactions is a complex task which relies on a coordinated set of processes that allow people to make sense of their social environments through understanding their own emotions, as well as the emotions and interpersonal motivations of others. Referred to as social cognition (Fiske & Taylor, 1991), this set of processes encompasses other constructs including perceptual biases, TOM/empathy, and social problem-solving. Inherently related to social cognition is the ability to correctly identify the emotions of others. In fact, emotion recognition is central to the ability to accurately perceive, interpret, and respond appropriately to social cues. Studies examining emotion recognition in BPD highlight how difficulties in interpersonal relationships may be related to less accurate detection and identification of facial affect in others (Bland, Williams, Scharer, & Manning, 2004; Levine, Marziali, & Hood, 1997; Merkl et al., 2010; Unoka, Fogd, Füzy, & Csukly, 2011). Indeed, several research groups found that individuals with BPD were less accurate in recognizing facial expressions of emotion displayed at full intensity, either neutral or negative, compared to healthy controls (e.g., Levine et al., 1997). Others have pointed out that the impairment seems to be limited to recognition and discrimination among negative emotions (Bland et al., 2004; Guitart-Masip et al., 2009), especially fear and anger (Unoka et al., 2011). Those with BPD may be more likely to misinterpret neutral expressions as negative (Dyck et al., 2009) and have poorer recognition of integrated facial/prosodic emotions when they are displayed together (Minzenberg, Poole, & Vinogradov, 2006). However, the results on this topic are mixed. Data from some studies suggest no impairment in emotion recognition in individuals with BPD or BPD symptoms compared to healthy controls (Domes et al., 2008; Gardner, Qualter, Stylianou, & Robinson, 2010) and there is some evidence of faster accurate responding in those with BPD compared to controls when detecting emotion in faces morphing from neutral to full expressions (Lynch et al., 2006). These mixed findings are difficult to synthesize due to the different methodologies, facial affective stimuli, the possible influence of psychotropic medications, and co-occurring psychiatric problems across studies. For example, research suggests that those with mood

196

S.A. Lazarus et al. / Clinical Psychology Review 34 (2014) 193–205

(e.g., Bistricky, Ingram, & Atchley, 2011) and anxiety (e.g., Shin et al., 2005) disorders demonstrate dysfunctional processing of facial affective cues. Given the high rates of co-occurring mood and anxiety disorders (e.g., 96.3% and 88.4%, respectively; Zanarini et al., 1998) among those with BPD, it is extremely difficult to determine whether deficits in facial affect recognition characterize BPD, specifically, or are better accounted for by comorbid symptom disorders. In addition, there is evidence that anti-depressants enhance selective attention towards positively valenced stimuli in depressed adults (Wells, Clerkin, Ellis, & Geevers, 2014), and anxiolytic interventions for social anxiety attenuate amygdala reactivity to angry and fearful facial expressions (Phan et al., 2013). It is possible that use of psychotropic medications by those with BPD may influence facial affect recognition and the direction of the influence (e.g., faster/ slower response, increased/decreased attention to threat) may depend on a complex interaction among BPD status, medication regimen, and specific comorbid symptom disorders. Accordingly, it is important to account for current psychotropic medication use and comorbid disorders when attempting to reconcile contradictory findings in the area of facial affect recognition. Another possible explanation for the discrepancies in findings, suggested by Daros, Zakzanis, and Ruocco (2013), is related to an interaction between the intensity of the displayed facial emotion and baseline emotional arousal. For example, when viewing faces of lower intensity, high baseline emotional arousal may enhance facial emotion detection abilities, whereas when emotions are presented at full intensity, the same baseline arousal may actually interfere with accurate perception. This would help explain earlier detection of facial emotion (Lynch et al., 2006) and also impairment at full presentation that is generally specific to discrimination among negative emotions. Studies that examine the ability to decode mental states from pictures of only the eye area (Reading the Mind in the Eyes Test, RMET; Baron-Cohen, Wheelwright, Hill, Raste, & Plumb, 2001) in individuals with BPD also yield somewhat mixed findings. While in two studies (Fertuck et al., 2009; Scott, Levy, Adams, & Stevenson, 2011) BPD was related to more accurate perception of mental states, another study (Preißler, Dziobek, Ritter, Heekeren, & Roepke, 2010) found no difference between women with BPD and healthy controls. Fertuck et al. (2009) found that averaged across conditions, the BPD group was superior to the healthy control group on the mental state discrimination task; when the conditions were examined separately, the BPD group advantage was evident for positive and neutral stimuli, but not for negative stimuli. However, when controlling for severity of depressive symptoms, participants with BPD were superior to healthy controls only on the total score of RMET, suggesting that inclusion of depressive symptoms weakens the effect of BPD status on RMET performance. Scott et al. (2011) came to the same conclusion as Fertuck et al. (2009) but found the opposite pattern at the level of emotion valence; the high BPD feature group responded more accurately than the low BPD feature group but only for negatively valenced stimuli. Taking these two studies together, it is possible that the interaction between baseline arousal and intensity of the emotional stimuli may again help to explain the inconsistencies. If individuals with BPD features (i.e., Scott et al., 2011) have lower levels of baseline emotional arousal than those with diagnosed BPD (i.e., Fertuck et al., 2009) and positive or neutral emotions can be assumed to be less emotionally arousing for those with BPD and related features, Daros et al.'s (2013) hypothesis might explain the superior performance of the BPD feature group in the negative valence condition and the BPD disorder group in the positive and neutral conditions. The results from these studies indicate that BPD is not related to any clear deficits in the perception of mental states from just the eye region, and may actually be associated with increased accuracy for particular stimuli. Replication of these results with both clinical and subclinical samples would help clarify the impact of valence and severity of BPD symptoms on mental state perception. Findings of less accurate facial recognition may suggest that individuals with BPD are distracted by

increasing amounts of information provided by the entire facial region, which is consistent with findings that those with BPD perform worse when decoding combined facial and prosodic information (Minzenberg et al., 2006). It is also possible that an increase in emotional information leads to higher arousal, which may moderate the effect of increased complexity of stimuli on accuracy. Given that we typically have an abundance of facial and extralinguistic information in day-to-day life, it is important to continue this research examining whether individuals with BPD may face a disadvantage in real-world social interactions. 2.1.1. Perceptual biases Related to facial emotion perception in BPD is the issue of whether those with BPD have more negative and extreme views of others compared to healthy individuals and those with other personality disorders (PDs). In order to examine perceptual biases in BPD, in several studies (Arntz & Veen, 2001; Sieswerda, Barnow, Verheul, & Arntz, 2013; Veen & Arntz, 2000) Arntz and colleagues asked participants to evaluate actors in film clips to assess whether BPD is related to more extreme or negative ratings of others. Veen and Arntz (2000) found that when accounting for the contribution of other Axis II disorders and selfreported emotional responses to clips, individuals with BPD had more extreme categorizations of others (i.e., responses further from zero on a visual analog scale of opposing qualities) for situations with BPDspecific themes (e.g., rejection, abandonment) than individuals with Cluster C PDs or healthy controls. However, Sieswerda et al. (2013) found no evidence of more extreme categorization of others by individuals with BPD compared to controls. Taking these studies together, individuals with BPD and Cluster C PDs had stronger self-reported emotional responses to films (Veen & Arntz, 2000) and provided less complex evaluations of the actors in the clips (Arntz & Veen, 2001). Individuals with BPD also had a tendency to rate actors more negatively than healthy controls (Sieswerda et al., 2013) and those with Cluster C PDs (Arntz & Veen, 2001). There are at least two potential explanations for the discrepancy in the dichotomous thinking findings. First, the control group in the Sieswerda et al. (2013) study had higher dichotomous thinking scores than the control group in the Veen and Arntz (2000) study; thus, it is possible that the control group in the Sieswerda et al. (2013) study had more problematic interpersonal processing than would be representative of “healthy controls” in general. Second, Sieswerda et al. (2013) found that severity of childhood sexual abuse was associated with dichotomous thinking and this association was not tested in the Veen and Arntz (2000) study. This highlights the importance of including measures of potential moderators of the relationship between BPD and interpersonal functioning in order to determine for whom and under what conditions interpersonal difficulties may arise. Performance-based projective measures of object relations have been used to assess possible biases in how those with BPD view others. On these measures, individuals with BPD have more malevolent representations of others, lower capacity to invest emotionally in relationships (Segal, Westen, Lohr, Silk, & Cohen, 1992), and less mature/ balanced investment in relationships than those with MDD and healthy controls (Westen, Lohr, Silk, Gold, & Kerber, 1990). Further, among individuals with BPD, those with non-suicidal self-injury (NSSI) had more negative expectations of treatment from others and poorer understanding of social causality of interpersonal interactions than those without concomitant NSSI (Whipple & Fowler, 2011). Together, these results indicate that individuals with BPD are more negative in the way they view others and in their expectations for relationships than controls. Further, more severe BPD symptoms, as indicated by the presence of NSSI, may be related to more extreme negative expectations in relationships. In contrast to these studies, Tragesser, Lippman, Trull, and Barrett (2008), using a vignette design, found that BPD features were not associated with more negative perceptions of a “teaser” or more negative attributions for teasing behavior. This is despite the fact that BPD features were related to the predicted experience of more negative affect in

S.A. Lazarus et al. / Clinical Psychology Review 34 (2014) 193–205

response to imaginal teasing. In light of the clear differences in methodologies used, it is difficult to draw conclusions about the diverging pattern of findings between this study and the studies that reported perceptual biases in BPD. One explanation for the inconsistency is that reading about a hypothetical situation is less emotionally evocative than viewing video clips, as in the Arntz and Veen (2001), Veen and Arntz (2000), and Sieswerda et al. (2013) studies. It is also possible that individuals with BPD features perceive teasing less negatively than other interpersonal behaviors (e.g., abandonment). Additionally, the participants in the Tragesser et al. (2008) study had elevated BPD features as opposed to diagnoses of BPD (as in the other studies). It is possible that perceptual biases are more pronounced among individuals meeting full criteria for BPD than in college students with subthreshold BPD features. 2.1.2. Theory of Mind (TOM) and empathy Researchers have also examined whether individuals with BPD have impairments related to understanding social cues, including reactions of others. Specifically, it is posited that individuals with BPD have deficits in TOM abilities, or one's ability to understand and correctly interpret the mental states that underlie other people's observable behaviors (Fonagy & Target, 1996), which interfere with successful social interactions. Several studies have been conducted that assess participants' ability to understand the mental states of others with stories and movie clips depicting social interactions. For example, Arntz, Bernstein, Oorschot, and Schobre (2009) found no evidence of TOM deficits in individuals with BPD when inferring the thoughts, intentions, and feelings of characters in standardized stories of nuanced interpersonal behaviors (e.g., white lies, persuasion, and sarcasm). Harari, Shamay-Tsoory, Ravid, and Levkovitz (2010), however, found that individuals with BPD performed worse than controls on cognitive (i.e., ability to detect the speaker's state of knowledge), but not affective (i.e., ability to understand the emotional impact of the statement on the listener) understanding of a social “faux pas.” Preißler et al. (2010) provided further evidence of TOM deficits in BPD using a task in which participants viewed a movie of people interacting in various social contexts, such as a dinner party, and throughout the video were asked to interpret the characters' mental states. When compared to healthy female controls, women with BPD showed poorer recognition of emotions, thoughts, and intentions of the characters. Thus, when using stories and movies, there is some evidence that individuals with BPD have impairments in TOM abilities. However, it may be important to differentiate between cognitive and affective abilities. Affective understanding may be more akin to empathy, while cognitive understanding involves an understanding of norms, intentions, and anticipated outcomes and is likely to reflect a more complex social understanding. Examining responses to pictures of other people experiencing emotions is another way to assess the ability to understand and empathize with the emotions of others. New et al. (2012) found no differences between individuals with BPD and healthy controls in their ratings of the feelings of people in positive, negative, and neutral interpersonal pictures. Dziobek et al. (2011) found deficits in both cognitive (i.e., infer mental state) and affective (i.e., rate level of empathic concern) empathy in those with BPD compared to healthy controls when shown pictures intended to produce strong emotional reactions (e.g., a child in a war scene). The authors also found some evidence of differences in neural activation while viewing emotional pictures. During the cognitive empathy component of the task, women with BPD, compared to female healthy controls, had significantly reduced response in the left superior sulcus and gyrus, a region which is thought to be related to the ability to think about others (Saxe & Kanwisher, 2003). During the affective empathy component of the task, women with BPD, compared to healthy controls, showed increased response in the right middle insular cortex, an area often associated with emotional reactions to others that are selforiented (Jackson, Brunet, Meltzoff, & Decety, 2006). Similarly, in a study by Mier et al. (2012), individuals with BPD and healthy controls completed a task assessing affective TOM during an event-related fMRI

197

design. Although there was no behavioral evidence of deficits in TOM in the BPD group, there were differences in neural activation between the groups. Whereas healthy controls showed increased activation in the mirror neuron system with increasing task complexity, the BPD group exhibited hypoactivation in these areas as well as increased activation in the left amygdala. Hypoactivation of areas related to the mirror neuron system could underlie a learned response to modulate strong emotions related to increased emotional sensitivity which may be reflected in the heightened amygdala reactivity. This suggests that individuals with BPD may rely on affect-dominated processing, which when combined with inadequate prefrontal control, may interfere with social processing, especially under conditions that are complex or involve high emotional arousal. These findings offer important insights into possible patterns of regional neural activation underlying deficits in empathy in BPD. However, it will be important to extend this research using additional control groups and diverse methods, such as those used to investigate neural connectivity, in order to more precisely characterize specific neural mechanisms involved in interpersonal impairments in BPD. The mixed findings for affective TOM may be related to the methodology of the studies which differ in several ways that may potentially impact performance. First, the stimuli differ in their valence and arousal levels. For example, pictures chosen to be particularly emotionally evocative (Dziobek et al., 2011) likely elicit differences between those with BPD and controls, while standardized stories (Arntz et al., 2009) which are not chosen to be particularly emotional (but rather to depict social situations) may be less emotionally arousing and allow individuals with BPD to perform well. Second, it is likely that stimuli with BPD-specific content evoke stronger emotional reactions than more general stimuli. For example, it is possible that detecting the emotional impact of a social faux pas is especially relevant to concerns of abandonment and sensitivity to rejection in BPD and this interferes with task performance. This is opposed to other stimuli that may be less personally relevant for individuals with BPD, such as that used in Mier et al. (2012) which asks the participant to match a facial expression with an emotional intention. Finally, tasks that require more effort on the part of the participant and/or involve more cognitive load may be more emotionally arousing and more difficult for individuals with BPD. Consistent with this, methodology that assesses social understanding periodically throughout a movie (Preißler et al., 2010) is likely more challenging than a task that uses static pictures to assess this ability (Arntz et al., 2009; New et al., 2012). Given the high emotional sensitivity in individuals with BPD, these small differences in social stimuli may have a meaningful impact on performance. Other researchers have used behavioral laboratory tasks that simulate interacting with another partner to examine the ways in which individuals with BPD may differ from others in terms of social cognition. For example, Wischniewski and Brüne (2012) examined participants' perceptions of and responses to unfair offers in a version of the dictator game wherein individuals with BPD and clinical controls with no PD observed two “other players” (i.e., dictator and recipient played by the computer). Participants could punish the dictator for unfair offers by taking away money to give to the recipient. Individuals with BPD were no more (or less) likely to punish the dictator than participants without a PD; both groups increased punishment in accordance with the degree of unfairness of the offer, suggesting that those with BPD are as able to evaluate and respond to observed unfairness by others. Similarly, Franzen et al. (2011) used a virtual social exchange game to examine the influence of emotional cues and assessment of fairness on the ability to judge a partner's intentions. BPD and healthy control participants acted as the investor, deciding how much money to invest in a multi-round trust game with four different trustees, who varied both in the trustworthiness of their offers and emotional expressions (e.g., neutral or congruent with the fairness of their offer) throughout the round. Whereas for the BPD participants, the trustworthiness of the partner's

198

S.A. Lazarus et al. / Clinical Psychology Review 34 (2014) 193–205

offer was related to their investment regardless of emotional expression, for controls, the emotional cues were weighed more heavily than trustworthiness in investment decisions. This is despite the fact that the two groups performed comparably on an earlier emotion recognition task that used the trustees' faces as stimuli, suggesting that individuals with BPD did not use the emotional information in evaluating fairness of an offer, even though it was available to them. This is consistent with the findings of increased sensitivity or vigilance to cues of social threat, with participants with BPD attending to the unfair offers regardless of additional information provided by emotional expression. It also supports the presence of intact affective TOM in BPD, but differences or deficits in cognitive TOM. Other behavioral laboratory tasks that involve the correct sequencing of social scenarios also have been used to assess TOM abilities. For example, Ghiassi, Dimaggio, and Brüne (2010) examined the performance of individuals with BPD and healthy controls on sequencing of social scenarios (Brüne, 2005) depicting cooperation and cheating and found no differences between the two groups. However, Schaffer, Barak, and Rassovsky (2013) found that individuals with BPD scored lower on the Picture Arrangement subtest from the Wechsler Adult Intelligence Scale (WAIS-III; Wechsler, 1998) compared to controls and that performance did not depend on whether or not contextual information (i.e., a title for each scene) was provided. Context did, however, have an impact on response time. When contextual information was provided, the BPD group was slower to respond than controls (M = 27.3 s and 16.6 s, respectively); however, when contextual information was not provided, the BPD group responded more quickly than controls (M = 18.9 s and 24.9 s, respectively). Participants with BPD also performed worse than controls on a test of general social perception (i.e., the ability to label facial expressions, voice intonations, and bodily gestures containing social cues). This suggests that those with BPD may misinterpret social cues at relatively basic levels, which then may lead to difficulty integrating social cues and understanding social situations, resulting in further interpersonal difficulties. Although it did not impact accuracy, contextual information actually impaired the performance of those with BPD by increasing reaction time to complete the social perception task. This may suggest that individuals with BPD are overwhelmed by contextual interpersonal cues, even when such cues are designed to facilitate responding (and do so for those without BPD). Given the deficit in accurately labeling interpersonal cues such as tone of voice and gestures, contextual cues may not serve the anticipated facilitative functions for those with BPD. The differential impact of contextual information on BPD versus control participants, as well as the neuroimaging results from Mier et al. (2012), suggest that although some important interpersonal information is equally available to individuals with BPD, this information may be processed or used less effectively. Consistent with this, Franzen et al. (2011) found that although individuals with BPD were equally likely to recognize the emotion of their interactional partners, they did not use this information in the same way as control participants when evaluating the trustworthiness of an offer.

problem-solving means (Kehrer & Linehan, 1996; Kremers, Spinhoven, Van der Does, & Van Dyck, 2006) and individuals with BPD provide responses on the MEPS that are less relevant to the problem (Maurex et al., 2010) compared to healthy controls. Further, when compared to healthy controls and those with other Axis I and Axis II disorders, individuals with BPD and clinical controls provided solutions that were less relevant, effective, and specific than healthy controls (Bray, Barrowclough, & Lobban, 2007). While impairments in both patient groups may suggest that the deficits are related to negative affect or psychopathology in general, there is evidence that those with BPD provided responses that were less specific than both clinical and healthy controls. Further, in a sample of individuals with BPD who were chronically suicidal, Kehrer and Linehan (1996) found that inappropriate means on the MEPS (i.e., substance abuse, lying, parasuicidal behavior) at four and eight months significantly predicted subsequent NSSI (i.e., parasuicidal) behaviors. These studies consistently find worse performance on the MEPS for those with BPD than controls. In addition, while those with other PDs may share some deficits in social problem-solving with those who suffer from BPD, there also appear to be impairments that are unique to BPD, such as reduced specificity of means. Because difficulties with social problem-solving are related to serious outcomes in BPD, such as NSSI, further research is needed to clarify whether quality of social problem-solving is differentially impaired among individuals with BPD compared to other clinical populations. Using the observation that many problematic behaviors associated with BPD occur in the context of emotional arousal, Dixon-Gordon, Chapman, Lovasz, and Walters (2011) explored the hypothesis that a negative emotion induction may interfere with the problem-solving abilities of individuals with BPD features. Their results suggest that emotional context is an important factor to consider when evaluating the relation between social problem-solving skills and BPD. Following a negative emotion induction (i.e., imagining rejection from a romantic partner and then from friends while seeking support), the high BPD features group generated fewer relevant solutions and more inappropriate solutions to social problems than the low BPD features group. In addition, increases in self-reported negative emotions in response to the rejection stressor mediated the relationship between BPD features and reduced social problem-solving. Thus, the results from several studies using the MEPS indicate impaired social problem-solving in BPD. This suggests that deficits in interpersonal functioning are at least partially related to trouble coming up with specific, active solutions when faced with difficult interpersonal interactions. The study by Dixon-Gordon et al. (2011) highlights the importance of context, specifically emotional state, in the assessment of interpersonal functioning in those with BPD. As a next step in this research, it will be important to replicate these effects in a sample of individuals meeting diagnostic criteria for BPD, as opposed to heightened BPD features. As suggested when evaluating the facial affect recognition literature, it is possible that heightened baseline emotional arousal in BPD interferes with performance on social problem-solving tasks. 2.2. Reactivity to interpersonal stressors

2.1.3. Social problem-solving Another factor influencing interpersonal functioning is the ability to respond flexibly to common problems encountered during social interactions. This ability involves functions related to TOM (i.e., the ability to acquire an accurate understanding of social context) as well as problemsolving skills needed to develop appropriate solutions to interpersonal difficulties. The means-end problem-solving task (MEPS; Platt, Spivak, & Bloom, 1971) has been used in several studies as an assessment of social problem-solving in BPD. Participants are asked to determine how they would arrive at a given solution to an interpersonal problem when they are provided with the beginning and end of the problem scenario and responses are then coded on relevant dimensions such as appropriateness, activity, passivity, and focus on emotion regulation. The results of several studies suggest that BPD is associated with more passive social

As reviewed above, individuals with BPD may not consistently demonstrate a perceptual bias, but there is some evidence that they have stronger emotional reactions in the context of social interactions compared to others. In research examining reactivity to interpersonal stressors, vignettes describing interpersonal situations, behavioral tasks approximating social contexts, and actual interpersonal interactions are used to examine self-reported emotional reactions and biological correlates of these stressors. Tragesser et al. (2008) found that undergraduates with high BPD features reported that they would be more likely to feel both angry and sad in reaction to imagined teasing than those with low BPD features, regardless of the source (i.e., friend or stranger) or the content (i.e., sensitive or non-sensitive topic) of the teasing. However, in a laboratory paradigm involving negative social

S.A. Lazarus et al. / Clinical Psychology Review 34 (2014) 193–205

and academic feedback, Chapman, Walters, and Dixon-Gordon (2012) found that the content of the social feedback did influence emotional reactions. In a social feedback condition, but not an academic feedback condition, individuals with high BPD features showed a significant increase in negative emotions. Controls showed the opposite pattern (i.e., a significant increase in negative emotions to the academic stressor, but not the social stressor). This suggests that individuals with elevated BPD features may be especially emotionally sensitive to feedback when it involves self-relevant social information and may be less sensitive to feedback related to performance (even when it is equally self-relevant). While Tragesser et al. (2008) found no effect of the source of the feedback, the authors suggest that the study may not have had adequate power to detect these differences in their sample. Also, in the manipulation by Chapman et al. (2012), participants believed they were receiving feedback from an actual peer rather than imagining themselves in the scenario; it is likely that “actual” feedback is a more powerful stressor than imagined teasing. Taken together, these studies lend support to the conclusion that individuals with elevated BPD features may experience more unpleasant emotions (e.g., anger, sadness) in stressful interpersonal contexts compared to other contexts and/or individuals with low BPD features. In a series of studies examining the reactions of individuals with BPD to an interpersonal stressor, Staebler et al. (2011) and Renneberg et al. (2011) used Cyberball (Williams & Jarvis, 2006), a virtual ball toss game, to simulate social inclusion or exclusion. Participants (i.e., females with BPD and healthy controls) were told they were playing the ball toss game in order to practice their mental visualization skills with two other participants and were randomly assigned to either an inclusion or exclusion condition. Participants with BPD reported receiving the ball a lower percentage of the time in the inclusion condition than controls (Staebler et al., 2011) and felt more readily excluded regardless of condition (Renneberg et al., 2011). In terms of subjective emotional responses, in both studies individuals with BPD reported more negative emotions than controls overall, regardless of condition (i.e., inclusion or exclusion). In the Staebler et al. (2011) study, individuals with BPD reported greater self-focused negative emotions (e.g., sadness, loneliness) than controls both before and after playing Cyberball. After exclusion, those with BPD had an increase in other-focused emotions (e.g., resentment, anger) whereas controls did not. These results suggest that the interpersonal difficulties experienced by individuals with BPD may be related to a biased perception of inclusion or participation and an increased likelihood to experience negative emotions across social contexts, compared to controls. Further, it appears that negative other-focused emotions are particularly impacted by exclusion for individuals with BPD, while self-focused negative emotions tend to be higher in those with BPD than controls overall. Studies that examine the biological underpinnings of emotional reactivity point to several potential neurobiological correlates of stronger emotional responses to interpersonally relevant contexts and stimuli. For example, Walter et al. (2008) examined salivary cortisol levels in individuals with BPD and controls in response to a 10-minute discussion with their mothers of a conflict topic provided by the participants. Although participants with BPD did not differ from controls on baseline or peak cortisol levels, they showed a delayed recovery of cortisol response following the conflict discussions, supporting the theoretical assertion that individuals with BPD have a slower return to baseline after emotional arousal (Linehan, 1993). Similarly, using a public speaking social stressor, Simeon, Knutelska, Smith, Baker, and Hollander (2007) examined hypothalamic-pituitary-adrenal (HPA) axis activity (through cortisol and norepinephrine levels) as a function of dissociation, a criterion for BPD. When BPD participants with either high or low dissociative experiences were compared with healthy controls, the three groups differed significantly in peak cortisol reactivity, with the BPD high dissociation group having a more robust peak response than the BPD low dissociation and control groups; the BPD low dissociation group did not differ from controls. There were no differences in

199

norepinephrine stress reactivity. This suggests that dissociative processes in the context of public speaking among individuals with BPD may lead to or be the result of higher cortisol activity. Future studies are needed to replicate and extend these finding to other BPD-relevant interpersonal contexts. The findings of these two studies combined point to the possibility that individuals with BPD may have increased cortisol reactivity (perhaps primarily in the delayed return to baseline) in stressful interpersonal contexts. At least one study has used imaging technology to examine differences in response to interpersonal stimuli between those with and without BPD (i.e., healthy controls). Ruocco et al. (2010) used functional near infrared spectroscopy (FNIRS) to assess neural activity during a task that simulated social inclusion and exclusion via a card game played with confederates. In analyses focusing on the medial prefrontal cortex (mPFC), an area implicated in social–cognitive functions related to the self and interactions with others (Saxe, 2006), BPD and healthy control participants showed similar levels of oxygenated hemoglobin in the inclusion condition. When excluded from the card game for most of the scan, participants with BPD showed greater activation in the left mPFC than controls. This finding is consistent with previous research reporting that individuals with BPD have dysfunction in fronto-lymbic circuitry (New, Goodman, Triebwasser, & Siever, 2008). However, some studies find frontal hypoactivation in cognitive control areas when attempting to regulate emotional reactions to interpersonal stimuli in BPD (e.g., Koenigsberg et al., 2009). Thus, it appears that PFC control in response to emotional reactions, which likely plays a role in top-down regulation of emotion, often differs in those with BPD and healthy controls. Whether this difference is consistently reflected in increased or diminished reactivity or both increased and diminished reactivity in divergent contexts, however, remains unanswered. Further, the extent to which dysfunction in cognitive control areas, hyperactivity of limbic regions, or issues of connectivity among these areas play a role in interpersonal disruptions in BPD is still unclear. As newer technology (e.g., Magnetoencephalography; MEG) allows us to build in more fine-grained analyses of temporal responding, hopefully we can better understand this neurological “cascade.” These studies examining reactivity to interpersonal stressors in BPD collectively suggest that individuals with BPD may have different biological reactions to interpersonal situations, including greater cortisol response to social stressors and more neural activation in the left medial prefrontal areas in response to social exclusion, than control participants. Also, individuals with BPD have greater self-reported emotional reactions (i.e., more negative affect and anger) to interpersonal stressors compared both to non-interpersonal stressors and other participants. Individuals with BPD may be more likely to perceive exclusion and appear to have stronger negative emotions in contexts that highlight participation, regardless of inclusion or exclusion feedback. An important next step in this area of research is to replicate and extend the findings reviewed above using samples of individuals meeting full criteria for BPD and clinical control groups alongside healthy controls. Until such studies are conducted, it will not be clear whether the findings above can be considered conclusive and specific to BPD. 2.3. Interpersonal aggression Another area of impairment that has been investigated in BPD is interpersonal aggression. Although aggressive behaviors are widely considered to be central to BPD, few studies have examined aggression in an interpersonal context experimentally. As expected, BPD is associated with increased aggression in interpersonal contexts. In response to vignettes describing teasing scenarios, BPD features were related to wanting to engage in aggressive behaviors, such as glaring at someone or making retaliatory comments (Tragesser et al., 2008). BPD features were also related to ratings of the imagined likelihood of engaging in the aggressive behavior. Other research in this area utilizes a behavioral laboratory task that assesses aggressive responding (i.e., Point-

200

S.A. Lazarus et al. / Clinical Psychology Review 34 (2014) 193–205

Subtraction Aggression Paradigm; PSAP) in which participants believe they are playing a game against an opponent and can either accumulate money for themselves or subtract money from their opponents. On this task, individuals with BPD consistently respond more aggressively (i.e., subtracting from one's opponent) than healthy controls (Dougherty, Bjork, Huckabee, Moeller, & Swann, 1999; McCloskey et al., 2009). In a study by McCloskey et al. (2009), both BPD and other personality disorder (OPD) groups showed elevated levels of aggressive responding on the PSAP compared to the control group and did not differ from each other. However, self-reported aggression was related to aggressive responding in the BPD group but not in the OPD group, suggesting aggressive responding on the PSAP for this group may reflect something other than trait aggression. Alternatively, the OPD group may be less likely to see themselves as aggressive compared to those with BPD, despite displaying similar levels of behavioral aggression. New et al. (2009) examined the neural correlates of aggression using the PSAP among individuals with BPD who were high in aggressiveness, as evidenced by meeting criteria for intermittent explosive disorder (IED), and healthy controls. Participants with both BPD and IED were more aggressive than controls on the PSAP and also showed greater activation in areas of the brain associated with emotion, including the amygdala and orbitofrontal cortex, whereas healthy controls had increased activation in brain regions involved in top-down cognitive control, such as the dorsal prefrontal cortex. Thus, the findings from this study support previous research findings that individuals with BPD have increased aggressive responding, and add to this literature by suggesting that there is corresponding activation in regions of the brain associated with emotional processing and no increase in activation in emotional control centers during aggressive behavior. Overall, the studies investigating interpersonal aggression using a behavioral measure (PSAP) suggest that individuals with BPD may engage in more aggressive behavior in some experimentally manipulated interpersonal situations compared to healthy controls. The paradigms (e.g., being teased, monetary games) used in these studies suggest that in this context emotion may precede aggression to some degree but do not rule out the possibility that aggression precedes emotional response or that both are related to a third variable, such as impulsivity, in interpersonal interactions. Although this research supports the hypothesis that individuals with BPD tend to react more aggressively than others in interpersonal contexts, studies are needed that use clinical controls and occur in naturalistic settings to help advance an understanding of specific interpersonal contexts that differentially predict aggression among individuals with BPD compared to those with other psychiatric disorders. 2.4. Lack of cooperation/trust Recently, tasks from behavioral economics research involving trust and cooperation have been used to gain further insight into the interpersonal functioning of individuals with BPD. These tasks involve relying on others as well as understanding and responding to interpersonal cues. Unoka, Seres, Aspan, Bodi, and Keri (2009) found that individuals with BPD were less likely to invest money (i.e., trust their “partner”) than controls and depressed individuals, although they did not differ in their risk taking behavior. Further, those with BPD predicted worse outcomes for the trust game than controls and while controls increased the amount of money invested across trials there was no such trend for the BPD group. Interpersonal and cognitive symptoms of BPD accounted for 33% of the variance in the total amount of money invested in the trust game, leading the authors to suggest that decreased trust and the resultant behavioral performance in the game may be related to symptoms such as stress-related paranoia, identity disturbance, dissociation, and problems in interpersonal relationships. In addition, individuals with BPD appear to be less skilled at cooperation and behavior aimed at repairing ruptures than controls. In a multi-

round economic exchange game, King-Casas et al. (2008) observed that dyads with a BPD trustee showed a downward shift in investment levels in late rounds of the game, reflecting a break in cooperation. Healthy trustees were twice as likely as BPD trustees to coax (show generous gestures) in the presence of low offers, reflecting an attempt to enhance cooperation. During participation, when compared to controls, individuals with BPD exhibited differential responding in the anterior insula, an area known to respond to norm violations in various paradigms. Neuroimaging results indicated that whereas healthy controls showed strong linear responses in anterior insula activity to both offers from partners and money they repaid, BPD patients' insula activity was only related to the amount of money offered, not the amount repaid. Whereas the behavioral findings suggest that individuals with BPD are not attempting to repair breaks in cooperation, the neuroimaging findings indicate that these results require a more precise characterization. Specifically, the impairment for individuals with BPD may not be the ability to repair ruptures in cooperation, but a failure to perceive low offers from partners as violations of social norms, and, hence, a failure to see the need to repair the interpersonal transgression. This study highlights the importance of using biological measures in addition to behavioral assessments. When individuals with BPD demonstrate behavioral deficits compared to controls, there may be biological correlates of the behavioral deficits that point to potential mechanisms. Alternatively, when deficits are evident according to biological but not behavioral measures, we may suspect that individuals with BPD and controls are arriving at similar observable performances by divergent processes. One potential biological correlate of these apparent deficits in trust and cooperation may be oxytocin, a neuropeptide posited to play a central role in prosocial behavior (Carter, Williams, Witt, & Insel, 1992) and perceived trustworthiness of faces in humans (Theodoridou, Rowe, Penton-Voak, & Rogers, 2009). However, given that Oxytocin is also thought to increase the salience of social cues, Bartz et al. (2010) hypothesized that it might heighten concerns of rejection and abandonment in individuals with BPD, negatively impacting trust and cooperation. In this study, participants with BPD and healthy controls received intranasal oxytocin or placebo and played the assurance game (developed by Kollock (1998)), in which participants choose to cooperate or defect at the same time as their partner throughout three consecutive rounds. For BPD participants, oxytocin did not have its usual trustfacilitating effects. Compared to those receiving placebo, BPD participants receiving Oxytocin predicted less cooperation from partners and were more likely to defect in response to hypothetical partner cooperation, whereas controls followed the expected (i.e., opposite) pattern (expectation of more cooperation and increased likelihood of cooperation in response to hypothetical partner cooperation when receiving oxytocin compared to placebo). These findings support the authors' hypothesis that preexisting relationship expectancies may moderate the influence of oxytocin on prosocial behavior. In sum, it appears that individuals with BPD may be less likely than controls to cooperate in experimentally manipulated interpersonal contexts. Specifically, breakdowns in cooperation appear to be related to reduced trust in interactional partners as opposed to impulsivity in interpersonal situations (Unoka et al., 2009) or external reward associated with cooperation (Bartz et al., 2010). Further, there is preliminary evidence that oxytocin, known to enhance prosocial behavior, may have paradoxical effects on individuals with BPD. It is important to follow up this line of research to determine how personality constructs (e.g., attachment, avoidance), biological variables (e.g., oxytocin levels, amygdala activity), and interpersonal behaviors (e.g., cooperation, trust) interact to predict changes in trust and cooperation for those with BPD. 2.5. Behavior within mother–child interactions Researchers recently have begun to use behavioral measures of mother–infant interactions within a laboratory setting to investigate

S.A. Lazarus et al. / Clinical Psychology Review 34 (2014) 193–205

interpersonal functioning in BPD. Several studies have examined how mothers with BPD differ from ‘typical’ mothers by coding maternal behavior in interactions with their infants. These laboratory interactions include periods of face-to-face play and also situations that may be stressful to the infant, such as periods when the mother adopts a “still-face.” Overall, mothers with BPD tend to behave more intrusively (i.e., behavior that overwhelms or interferes with the child's efforts) and less sensitively (i.e., behavior that is not responsive to the child's needs), and are less competent at structuring their children's activities than control mothers (Crandell, Patrick, & Hobson, 2003; Hobson, Patrick, Crandell, García-Pérez, & Lee, 2005; Newman, Stevenson, Bergman, & Boyce, 2007). One laboratory procedure that has been used extensively to examine the behavior of mothers and their infants, the strange situation procedure (Ainsworth, Blehar, Waters, & Wall, 1978), has also been utilized with mothers with BPD to assess behavior during separation–reunion episodes. Using this methodology, Hobson et al. (2009) found that a higher proportion of mothers with BPD (85%) was classified as showing disrupted communication (e.g., mixed affective signals, inadequate responding to the infant's needs) than mothers with MDD (47%) and control (42%) mothers. Mothers with BPD also showed higher rates of frightened/disoriented behavior compared to mothers with MDD or healthy control mothers. In addition to the increased rates of problematic interactions with their children, mothers with high BPD pathology also are less likely than those with low BPD pathology to display positive affect in response to infant distress (Kiel, Gratz, Moore, Latzman, & Tull, 2011) and engage in fewer affiliative behaviors than control mothers (e.g., smiling and imitating) during typical interactions (White, Flanagan, Martin, & Silvermann, 2011). Specifically, White et al. (2011) observed the interactions of mothers with BPD, MDD, BPD + MDD, and healthy controls with their infants (mean age = 3.5 months). During the interactions, the mothers were told to interact with their infant as they would at home. Control mothers smiled more than mothers with MDD, BPD, and BPD + MDD whereas mothers with MDD smiled significantly more than mothers with BPD and BPD + MDD. In addition, mothers with MDD touched their infants more than mothers with BPD and BPD + MDD. In terms of game playing, both control and MDD mothers engaged in more game playing than BPD and BPD + MDD mothers, but they did not differ from each other. Lastly, mothers with BPD displayed less imitation with their infants than control mothers, mothers with MDD, or mothers with BPD + MDD. Although mothers with MDD were less responsive than controls in some behaviors, such as smiling, mothers with BPD (with or without MDD) showed the most consistent impairments. Kiel et al. (2011) examined the transactional nature of mother–child interactions; they reported that insensitive parenting by those with significant BPD pathology increased as infant distress persisted. For all dyads, infant distress was sensitive to maternal responsiveness, decreasing after positive affect and increasing after insensitivity by mothers. This mutual influence within the dyad underscores the importance of considering the transactional nature of mother–infant interactions, and interpersonal interactions more generally. Taken together, these studies suggest that mothers with BPD tend to be less competent in interacting with their infants, displaying deficits in areas such as sensitivity, structuring, and coping with infant distress. 3. Discussion The research reviewed here suggests several key areas of interpersonal impairment among individuals with BPD. First, consistent with object relations and cognitive theories of the disorder, individuals with BPD tend to view others more negatively and have more negative expectations for relationships than healthy controls. However, given that Tragesser et al. (2008) failed to find that BPD features were related to more negative views of an imagined “teaser,” it is unclear in which

201

contexts a negative bias towards others may be most likely to occur. Additional research is needed to characterize negative interpersonal interpretations and conclusions made by those with BPD using contextual and BPD-relevant cues, including emotionally evocative contexts. Together, research conducted in both laboratory and naturalistic settings will be needed to delineate the specific problematic ways in which individuals with BPD respond to others in interpersonal contexts. There are an increasing number of studies, with diverse methodologies ranging from behavioral economics paradigms to neuroimaging technology, that examine the ability of individuals with BPD to understand and interpret the mental states of others. The findings from these studies are somewhat mixed. In a number of studies, individuals with BPD do not differ from healthy controls in TOM skills and are able to appropriately understand and respond to social cues. However, there is some evidence to support the contention made by object relations theorists that those with BPD are less skilled at inferring the mental states of others and may struggle to apply and utilize these abilities across contexts. Harari et al. (2010) suggest one possibility for synthesizing these findings; namely that affective empathy is intact and cognitive empathy/understanding is impaired. In other words, individuals with BPD may be accurately sensing and connecting to the emotions others are feeling while simultaneously having difficulty understanding or cognitively contextualizing the emotions and thoughts of others. This is supported by the neuroimaging findings from Mier et al. (2012), which suggest that whereas amygdala hyperactivation may reflect exaggerated or increased sensitivity to affective perception of others' emotions, inadequate regulation through frontal regions may lead to impaired cognitive understanding in complex or emotional contexts. Further, this combination of hyperactive emotional responding (i.e., experience of emotion) combined with hypoactive emotion regulation (i.e., changing the course of emotion) may result in what appears to be erratic and/or impulsive behavior. Our understanding of this complex relationship would be furthered by research examining cognitive and emotional processing, as mechanisms of interpersonal behavior, across various levels of analysis (e.g., brain, behavior). The recent advent of the National Institute for Mental Health's (NIMH) Research Domain Criteria (RDoC; http://www.nimh. nih.gov/research-priorities/rdoc/index.shtml) may prove useful for future research in this area. For example, the RDoC may help provide direction to researchers seeking to better understand the underlying processes contributing to interpersonal impairments relevant, but perhaps non-specific, to BPD. Including multiple levels of analysis should help to contribute to the most fine-grained characterization. Finally, systematically manipulating contextual variables that may serve as moderators of a BPD–TOM relationship (e.g., relationship/ history with the interpersonal partner, emotional load, contextual cues) may help clarify the extent and pervasiveness of TOM deficits. One replicated result in the literature is that individuals with BPD tend to demonstrate impairment in social problem-solving skills compared to others. Multiple findings indicate that when attempting to solve interpersonally-relevant problems, individuals with BPD have a tendency to produce less effective means and more passive solutions. In addition, there is some evidence consistent with the biosocial theory that social problem-solving abilities are more adversely impacted by negative affect for individuals with BPD features than others. In the future, the role of emotion may be explicated by examining social problem-solving in more externally valid contexts, such as actual interpersonal conflicts in the laboratory. This is important given that in behavioral tasks that approximate interpersonal situations, individuals with BPD report stronger subjective emotional reactions to stressful social situations than other individuals (Tragesser et al., 2008), and respond more strongly to social stressors than other types of stressors (Chapman et al., 2012). Overall, the research on social cognition is complex, but has the potential to provide valuable insight into the interpersonal difficulties of those with BPD. It is important to understand and incorporate

202

S.A. Lazarus et al. / Clinical Psychology Review 34 (2014) 193–205

potential deficits in social cognition when evaluating findings from other domains. For example, the ability to accurately perceive facial emotion and understand the intentions and thoughts of others is an important part of evaluating the trustworthiness of a potential partner. More careful attention should be paid to alternative explanations, such as differences in social cognition and affect recognition, when interpreting differences between individuals with BPD and others in interpersonal behavioral tasks (such as cooperation tasks). Because individuals with BPD may experience deficits in several areas of interpersonal functioning, increasing the number of interpersonal domains assessed within a given study will allow for direct comparisons of areas of difficulty. Across the studies reviewed here, there is evidence of biological differences between individuals with BPD and others when processing interpersonal stimuli. First, increased emotional reactivity to interpersonal stimuli and contexts may be reflected in increased activation in the amygdala (Mier et al., 2012; New et al., 2009), increased cortisol reactivity (slower return to baseline than controls; Walter et al., 2008), and heightened activation in the mPFC (Ruocco et al., 2010). Second, there may be decreased activation in areas related to the ability to think about others (i.e., left superior sulcus and gyrus) and increased activation in areas related to self-oriented emotional reactions to others (i.e., right middle insular cortex; Dziobek et al., 2011). The increase in activation in the right middle insular cortex may reflect something akin to affective empathy; however, it is possible that the increased activation also leads to over-arousal in individuals with BPD. Neuroimaging findings by Mier et al. (2012) provide some support for the overarousal hypothesis. The authors found that while controls showed activation in the mirror neuron system during a TOM task, those with BPD actually showed hypoactivation. This may reflect implicit biological strategies to regulate strong emotional reactions. Further, the neural hypoactivation was not accompanied by any behavioral evidence of TOM deficits in those with BPD. Biological assessments of interpersonal functioning allow for identification of mechanisms related to behavioral differences between those with BPD and others and also provide information about different biological pathways that may lead to the same behavioral outcomes. Thus, while biological methods may provide means to test hypotheses about the mechanisms that underlie certain behavioral differences between those with BPD and controls, they may also suggest that different processes are at work for those with BPD and controls even when there are no differences in behavioral findings. As suggested by Daros et al. (2013) and consistent with the biosocial theory, baseline emotional arousal may explain some of the discrepant findings in studies of interpersonal functioning for those with BPD. Although the authors discuss their hypothesis within the domain of facial affect recognition, it may be applicable to understanding inconsistent findings in other areas as well. As suggested above, the discrepant findings in behavioral tasks assessing TOM may be related to varying levels of BPD symptoms interacting with the valence and arousal level of the stimuli. However, these factors are rarely accounted for in studies of interpersonal behavior. Further, although it may be assumed that stimuli of negative valence will be more emotionally arousing for individuals with BPD, this is still unclear. Thus, in any methodology that employs social stimuli, it will be important to be aware of and account for the valence and arousal level of those stimuli. Further, given that individuals with BPD are known to experience affective instability and intense emotions (for review see Rosenthal et al. (2008)), it is likely important to include state measures of emotion in studies with interpersonal stressors to explicate long-standing abilities from temporary deficits due to emotional over-arousal. Although we may assume that BPD symptoms are generally related to higher emotional arousal, this may not be true for all individuals with BPD. Other symptoms of the disorder such as dissociation or biological processes aimed at downregulation of sympathetic activity may be related to lower baseline arousal in some individuals at some times. Without such considerations,

the role of emotional experiences and emotion regulation difficulties in problems of trust, cooperation, aggression, or other interpersonal behaviors may be overlooked. Although there are several domains of mixed findings within interpersonal functioning in those with BPD, a consensus appears to be developing that disrupted interpersonal functioning in BPD extends to maternal behavior. Across a range of interactions, including free play, separation and reunion, and following still-face periods, mothers with BPD behave less responsively towards their infants than mothers without BPD pathology. In addition, this research (e.g., Kiel et al., 2011) highlights the transactional nature of interactions between mothers and infants, showing that infant distress leads to more insensitive parenting, which then leads to more infant distress. It is of note that the problematic transactions between mothers with BPD and their children are consistent with hypotheses generated from each of the primary theories of BPD development. Laboratory-based interpersonal interaction between mother and infant may be a useful model for research examining similar types of interpersonal behavior with adult partners. Additionally, longitudinal research that follows the children of mothers with BPD may provide important information about the development of interpersonal and emotion regulation skills given different developmental experiences in these areas. The findings reviewed here can be integrated to create several testable models of interpersonal dysfunction in those with BPD. One such model, consistent with both object relations and biosocial theories, would suggest that early insensitive, intrusive, or erratic interactions (perhaps with mothers with BPD or BPD features) lead to an increased vulnerability to interpersonal dysfunction in BPD. Although these findings do not necessarily suggest that all individuals with BPD have parents with the disorder, some studies suggest that the interpersonal symptoms of BPD are heritable (Zanarini et al., 2004). The vulnerability to interpersonal dysfunction may be characterized by mistrust of others, hypervigilance to rejection or abandonment cues, and difficulty recognizing and understanding emotions and extralinguistic cues from others. These vulnerabilities may increase the likelihood that those with BPD will perceive threat or find themselves in more problematic interpersonal situations. When threat is perceived (whether accurate or not), emotional (e.g., anger, rejection), behavioral (e.g., aggression), and biological (e.g., slow return to baseline after cortisol reactivity, increased amygdala activation, diminished PFC control) changes occur which make self-regulation difficult. In addition, those with BPD are likely to be less able to generate active and effective solutions in interpersonal situations which, in the face of dysregulated emotions, may lead to ineffective behaviors such as impulsivity and aggression, which in turn may further increase interpersonal stressors. Finally, in a chronic state of heightened affect, the individual is likely to perceive additional social threats, thus creating a vicious cycle. These different domains of functioning have not been tested concurrently and the vast majority of the studies reviewed here do not include longitudinal data. As such, it is premature to draw conclusions about the temporal and transactional relations that characterize interpersonal dysfunction for those with BPD but hypotheses such as these offer directions for future research. 3.1. Future directions Given that interpersonal dysfunction is part of the diagnostic criteria of BPD, it is important for research on this topic to move beyond demonstrating that individuals with BPD report having more difficulties in interpersonal functioning than healthy controls. To optimally characterize interpersonal dysfunction in BPD, it may be useful for future studies to avoid tautological outcomes by using dependent measures of interpersonal functioning that do not directly overlap with any given diagnostic criterion for BPD or by removing relevant criteria when examining the relation between BPD and interpersonal outcomes. For example, the DSM-5 (APA, 2013) lists frantic efforts to avoid abandonment, patterns of unstable relationships, interpersonal impulsivity, and aggression as diagnostic

S.A. Lazarus et al. / Clinical Psychology Review 34 (2014) 193–205

criteria for BPD. As such, measures of interpersonal difficulty that are not confounded with these criteria will reduce problematic overlap between predictor and outcome variables. Alternatively, studies aiming to characterize interpersonal dysfunction in BPD could examine the contexts (e.g., emotional arousal, interpersonal rejection) under which specific interpersonal impairments are more or less likely to be elicited. Such an approach would help to more precisely characterize interpersonal dysfunction in BPD beyond acontextual diagnostic constructs (e.g., reactive aggression), and may help to refine theoretical models and inform novel approaches to treatment. Finally, as we emphasize here, using multiple modes of assessment may avoid this potential confound between BPD diagnosis and interpersonal functioning. Specifically, rather than using self-reported BPD symptoms to predict self-reported interpersonal functioning, multimodal assessments provide a more detailed understanding of behavioral and physiological correlates of interpersonal problems in BPD. Further, including multi-modal assessments of interpersonal constructs will increase confidence that the characterizations of interpersonal problems for those with BPD are valid and stable. The NIMH RDoC may provide a useful heuristic framework to influence research in this area. Areas such as trust and cooperation are good examples of domains that have discrete, observable social behaviors that lend themselves to examinations involving both behavioral and biological assessments. The literature to date, including behavioral, biological, and self-report data, suggests that trust and cooperation are impaired for individuals with BPD. Given this consistency, we suggest that it is now appropriate to extend these studies to more ecologically valid contexts. For example, the role of trust and cooperation may be examined in therapy relationships or in relationships with meaningful others at varying levels of closeness. Further, it will be important to use informant methods to assess interpersonal symptoms in BPD, as the impact of these problems on others may be an important part of the picture. In addition, the next generation of treatments for BPD may benefit from including interventions specifically targeting skills needed to enhance trust and cooperation. Another area that deserves attention is related to determining the degree to which particular interpersonal impairments are specific to BPD. BPD has particularly high rates of comorbidity (or covariance) with many other disorders (e.g., MDD, post-traumatic stress disorder, substance use disorders, avoidant PD, narcissistic PD). As such, it is difficult to parse apart interpersonal dysfunction that is specific to BPD from interpersonal dysfunction associated with difficult personality traits more broadly (e.g., Neuroticism, impulsivity) or with severity of psychopathology (which may be reflected, in part, by meeting criteria for several disorders simultaneously). Some authors of the papers reviewed here have made attempts to begin addressing these issues by including relevant clinical control groups (e.g., MDD, Cluster C PDs, BPD + MDD) or including measures of relevant personality traits (e.g., impulsivity) or problematic behaviors (e.g., NSSI). We have tried to highlight findings that are specific to individuals with BPD as opposed to findings that differentiate all clinical groups from healthy controls. Future researchers should include control groups that cover the span of diagnostic overlap with BPD from the most similar (e.g., narcissistic PD, antisocial PD) to more dissimilar (e.g., avoidant PD, MDD). Finally, characterizing interpersonal difficulties associated with BPD is complicated by the heterogeneity of BPD presentations. BPD is, by current definitions, an extremely heterogeneous disorder and two individuals could both meet diagnostic criteria for BPD while sharing only one diagnostic criterion. Additionally, unlike some disorders (e.g., MDD), there are no “necessary criteria” for meeting the diagnostic threshold in BPD. As such, some of the mixed interpersonal findings reviewed here may result from individuals with BPD who have quite dissimilar vulnerabilities and reactions to interpersonal stressors. In the future, researchers may want to include samples that meet criteria for BPD and also share a common symptom cluster. For example, New et al. (2009) recruited participants who met criteria for both BPD and IED to assess aggressive responding. The selection of samples that

203

increase overlap of symptoms of interest may result in more stable and informative indicators of interpersonal difficulties. Further, the inclusion of potential moderators in studies of interpersonal functioning in BPD would likely aid characterization efforts by capitalizing on heterogeneity. In sum, interpersonal behaviors and relationships have long been theorized as central to the experience and expression of BPD. A large body of research speaks to the interpersonal distress and resultant consequences for individuals with BPD. More recently, researchers have turned their attention to the observable interpersonal behaviors that differentiate individuals with BPD from others in an attempt to understand how particular behaviors are related to interpersonal distress and to identify areas to target when working with patients with BPD. Based on the review of this literature, there appear to be replicable differences in the interpersonal behaviors of individuals with BPD or BPD features and those without such features, particularly in the areas of trust and cooperation and parenting behaviors. Future research that increases the ecological validity of assessments will further add to this knowledge base. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.cpr.2014.01.007. References Agrawal, H. R., Gunderson, J., Holmes, B.M., & Lyons-Ruth, K. (2004). Attachment studies with borderline patients: A review. Harvard Review of Psychiatry, 12, 94–104. http://dx.doi.org/10.1080/10673220490447218. Ainsworth, M., Blehar, M., Waters, E., & Wall, S. (1978). Patterns of attachment. Hillsdale, New Jersey: Erlbaum. American Psychiatric Association (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Arlington, VA: American Psychiatric Publishing. Arntz, A., Bernstein, D., Oorschot, M., & Schobre, P. (2009). Theory of mind in borderline and cluster-C personality disorder. The Journal of Nervous and Mental Disease, 197, 801–807. http://dx.doi.org/10.1097/NMD.0b013e3181be78fb. Arntz, A., & Veen, G. (2001). Evaluations of others by borderline patients. The Journal of Nervous and Mental Disease, 180, 513–521. http://dx.doi.org/10.1097/00005053200108000-00004. Baer, R. A., Peters, J. R., Eisenlohr-Moul, T. A., Geiger, P. J., & Sauer, S. E. (2012). Emotion-related cognitive processes in borderline personality disorder: A review of the empirical literature. Clinical Psychology Review, 32, 359–369. http://dx.doi.org/10.1016/j.cpr.2012.03.002. Baron-Cohen, S., Wheelwright, S., Hill, J., Raste, Y., & Plumb, I. (2001). The “Reading the mind in the eyes” test revised version: A study with normal adults and adults with Asperger syndrome or high functioning autism. Journal of Child Psychology and Psychiatry, 42, 241–251. http://dx.doi.org/10.1111/1469-7610.00715. Bartz, J., Simeon, D., Hamilton, H., Kim, S., Crystal, S., Braun, A., et al. (2010). Oxytocin can hinder trust and cooperation in borderline personality. Social Cognitive and Affective Neuroscience, 6, 556–563. http://dx.doi.org/10.1093/scan/nsq085. Bistricky, S. L., Ingram, R. E., & Atchley, R. A. (2011). Facial affect processing and depression susceptibility: Cognitive biases and cognitive neuroscience. Psychological Bulletin, 137, 998–1038. http://dx.doi.org/10.1037/a0025348. Bland, A.R., Williams, C. A., Scharer, K., & Manning, S. (2004). Emotion processing in borderline personality disorders. Issues in Mental Health Nursing, 25, 655–672. http://dx.doi.org/10.1080/01612840490486692. Bornstein, R. F., Becker-Matero, N., Winarick, D. J., & Reichman, A. L. (2010). Interpersonal dependency in borderline personality disorder: Clinical context and empirical evidence. Journal of Personality Disorders, 24, 109–127. http://dx.doi.org/10.1521/pedi.2010.24.1.109. Bouchard, S., Sabourin, S., Lussier, Y., & Villeneuve, E. (2009). Relationship quality and stability in couples when one partner suffers from borderline personality disorder. Journal of Marital & Family Therapy, 35, 446–455. http://dx.doi.org/ 10.1111/j.1752-0606.2009.00151.x. Bowlby, J. (1973). Attachment and loss: Vol. 2: Separation, anxiety, and anger. New York: Basic Books. Bray, S., Barrowclough, C., & Lobban, F. (2007). The social problem-solving abilities of people with borderline personality disorder. Behaviour Research and Therapy, 45, 1409–1417. http://dx.doi.org/10.1016/j.brat.2006.06.011. Brodsky, B.S., Groves, S. A., Oquendo, M.A., Mann, J. J., & Stanley, B. (2006). Interpersonal precipitants and suicide attempts in borderline personality disorder. Suicide and Life-Threatening Behavior, 36, 313–322. http://dx.doi.org/10.1521/suli.2006.36.3.313. Brown, M. Z., Comtois, K. A., & Linehan, M. M. (2002). Reasons for suicide attempts and nonsuicidal self-injury in women with borderline personality disorder. Journal of Abnormal Psychology, 111, 198–202. http://dx.doi.org/10.1037//0021-843X.111.1.198. Brüne, M. (2005). Emotion recognition, “theory of mind” and social behaviour in schizophrenia. Psychiatry Research, 133, 135–147. http://dx.doi.org/10.1016/j.psychres.2004.10.007.

204

S.A. Lazarus et al. / Clinical Psychology Review 34 (2014) 193–205

Carter, C. S., Williams, J. R., Witt, D.M., & Insel, T. R. (1992). Oxytocin and social bonding. Annals of the New York Academy of Sciences, 652, 204–211. Chapman, A. L., Walters, K. N., & Dixon-Gordon, K. L. (2012). Emotional reactivity to social rejection and negative evaluation among persons with borderline personality features. Journal of Personality Disorders. http://dx.doi.org/10.1521/pedi_2012_26_068 (Advanced online publication). Choi-Kain, L. W., Zanarini, M. C., Frankenburg, F. R., Fitzmaurice, G. M., & Reich, D. B. (2010). A longitudinal study of the 10-year course of interpersonal features in borderline personality disorder. Journal of Personality Disorders, 24, 365–376. http://dx.doi.org/10.1521/pedi.2010.24.3.365. Clarkin, J. F., Lenzenweger, M. F., Yeomans, F., Levy, K. N., & Kernberg, O. F. (2007). An object relations model of borderline personality disorder. Journal of Personality Disorders, 21, 474–499. http://dx.doi.org/10.1521/pedi.2007.21.5.474. Crandell, L. E., Patrick, M. P. H., & Hobson, R. P. (2003). “Still-face” interactions between mothers with borderline personality disorder and their 2-month-old infants. British Journal of Psychiatry, 183, 239–247. http://dx.doi.org/10.1192/bjp.183.3.239. Daros, A.R., Zakzanis, K. K., & Ruocco, A.C. (2013). Facial emotion recognition in borderline personality disorder. Psychological Medicine, 43, 1953–1963. http://dx.doi.org/10.1017/S0033291712002607. Dixon-Gordon, K. L., Chapman, A. L., Lovasz, N., & Walters, K. (2011). Too upset to think: The interplay of borderline personality features, negative emotions, and social problem solving in the laboratory. Personality Disorders: Theory, Research, and Treatment, 2, 1–18. http://dx.doi.org/10.1037/a0021799. Domes, G., Czieschnek, D., Weidler, F., Berger, C., Fast, K., & Herpertz, S.C. (2008). Recognition of facial affect in borderline personality disorder. Journal of Personality Disorders, 22, 135–147. http://dx.doi.org/10.1521/pedi.2008.22.2.135. Dougherty, D.M., Bjork, J. M., Huckabee, H. C. G., Moeller, F. G., & Swann, A. S. (1999). Laboratory measures of aggression and impulsivity in women with borderline personality disorder. Psychiatry Research, 85, 315–326. http://dx.doi.org/10.1016/S0165-1781(99)00011-6. Dyck, M., Habel, U., Slodczyk, J., Schlummer, J., Backes, V., Schneider, F., et al. (2009). Negative bias in fast emotion discrimination in borderline personality disorder. Psychological Medicine, 39, 855–864. http://dx.doi.org/10.1017/S0033291708004273. Dziobek, I., Preissler, S., Grozdanovic, Z., Heuser, I., Heekeren, H. R., & Roepke, S. (2011). Neuronal correlates of altered empathy and social cognition in borderline personality disorder. NeuroImage, 57, 539–548. http://dx.doi.org/10.1016/j.neuroimage.2011.05.005. Fertuck, E. A., Jekal, A., Song, I., Wyman, B., Morris, M. C., Wilson, S. T., et al. (2009). Enhanced ‘Reading the mind in the eyes’ in borderline personality disorder compared to healthy controls. Psychological Medicine, 39, 1979–1988. http://dx.doi.org/10.1017/S003329170900600X. Fiske, S. T., & Taylor, S. E. (1991). Social cognition (2nd ed.). New York, NY: England: Mcgraw-Hill Book Company. Fonagy, P., & Target, M. (1996). Playing with reality: I theory of mind and the normal development of psychic reality. International Journal of Psychoanalysis, 77, 217–233. Fonagy, P., Target, M., Gergley, G., Allen, J. G., & Bateman, A. W. (2003). The developmental roots of borderline personality disorder in early attachment relationships: A theory and some evidence. Psychoanalytic Inquiry, 23, 412–459. http://dx.doi.org/10.1080/07351692309349042. Franzen, N., Hagenhoff, M., Baer, N., Schmidt, A., Sammer, G., Gallhofer, B., et al. (2011). Superior “theory of mind” in borderline personality disorder: An analysis of interaction behavior in a virtual trust game. Psychiatry Research, 187, 224–233. http://dx.doi.org/10.1016/j.psychres.2010.11.012. Gardner, K. J., Qualter, P., Stylianou, M., & Robinson, A. J. (2010). Facial affect recognition in non-clinical adults with borderline personality disorder features: The role of effortful control and rejection sensitivity. Personality and Individual Differences, 49, 799–804. http://dx.doi.org/10.1016/j.paid.2010.07.018. Ghiassi, V., Dimaggio, G., & Brüne, M. (2010). Dysfunctions in understanding other minds in borderline personality disorder: A study using cartoon picture stories. Psychotherapy Research, 20, 657–667. http://dx.doi.org/10.1080/10503307.2010.501040. Guitart-Masip, M., Pascual, J. C., Carmona, S., Hoekzema, E., Bergé, D., Pérez, V., et al. (2009). Neural correlates of impaired emotional discrimination in borderline personality disorder: An fMRI study. Progress in Neuro-Psychopharmacology and Biological Psychiatry, 33, 1537–1545. http://dx.doi.org/10.1016/j.pnpbp.2009.08.022. Gunderson, J. G. (2007). Disturbed relationships as a phenotype for borderline personality disorder. The American Journal of Psychiatry, 164, 1637–1640. http://dx.doi.org/ 10.1176/appi.ajp.2007.07071125. Harari, H., Shamay-Tsoory, S. G., Ravid, M., & Levkovitz, Y. (2010). Double dissociation between cognitive and affective empathy in borderline personality disorder. Psychiatry Research, 175, 277–279. http://dx.doi.org/10.1016/j.psychres.2009.03.002. Hobson, R. P., Patrick, M., Crandell, L., García-Pérez, R., & Lee, A. (2005). Personal relatedness and attachment in infants of mothers with borderline personality disorder. Development and Psychopathology, 17, 329–347. http://dx.doi.org/10.1017/S0954579405050169. Hobson, R. P., Patrick, M. P. H., Hobson, J. A., Crandell, L., Bronfman, E., & Lyons-Ruth, K. (2009). How mothers with borderline personality disorder relate to their year-old infants. British Journal of Psychiatry, 195, 325–330. http://dx.doi.org/10.1192/ bjp.bp.108.060624. Jackson, P. L., Brunet, E., Meltzoff, A. N., & Decety, J. (2006). Empathy examined through the neural mechanisms involved in imagining how I feel versus how you feel pain. Neuropsychologia, 44, 752–761. Jacobson, E. (1964). The self and the object world. Oxford, England: International Universities Press. Kehrer, C. A., & Linehan, M. M. (1996). Interpersonal and emotional problem solving skills and parasuicide among women with borderline personality disorder. Journal of Personality Disorders, 10, 153–163. http://dx.doi.org/10.1521/pedi.1996.10.2.153. Kernberg, O. F. (1980). Internal world and external reality: Object relations theory applied. New York: Jason Aronson.

Kiel, E. J., Gratz, K. L., Moore, S. A., Latzman, R. D., & Tull, M. T. (2011). The impact of borderline personality pathology on mothers' responses to infant distress. Journal of Family Psychology, 25, 907–918. http://dx.doi.org/10.1037/a0025474. King-Casas, B., Sharp, C., Lomax-Bream, L., Lohrenz, T., Fonagy, P., & Montague, P. R. (2008). The rupture and repair of cooperation in borderline personality disorder. Science, 321, 806–810. http://dx.doi.org/10.1126/science.1156902. Klein, M. (1957). Envy and gratitude. New York: Basic Books. Koenigsberg, H. W., Fan, J., Ochsner, K. N., Liu, X., Guise, K. G., Pizzarello, S., et al. (2009). Neural correlates of the use of psychological distancing to regulate responses to negative social cues: A study of patients with borderline personality disorder. Biological Psychiatry, 66, 854–863. http://dx.doi.org/10.1016/j.biopsych.2009.06.010. Kollock, P. (1998). Social dilemmas: The anatomy of cooperation. Annual Review of Sociology, 24, 183–214. http://dx.doi.org/10.1146/annurev.soc.24.1.183. Kremers, I. P., Spinhoven, P., Van der Does, A. J. W., & Van Dyck, R. (2006). Social problem solving, autobiographical memory and future specificity in outpatients with borderline personality disorder. Clinical Psychology and Psychotherapy, 13, 131–137. http://dx.doi.org/10.1002/cpp.484. Levine, D., Marziali, E., & Hood, J. (1997). Emotion processing in borderline personality disorder. Journal of Nervous and Mental Disease, 185, 240–246. http://dx.doi.org/ 10.1097/00005053-199704000-00004. Linehan, M. (1993). Cognitive–behavioral treatment of borderline personality disorder. New York, NY: The Guilford Press. Links, P.S., & Heslegrave, R. J. (2000). Prospective studies of outcome: Understanding mechanisms of change in patients with borderline personality disorder. Psychiatric Clinics of North America, 23, 137–150. http://dx.doi.org/10.1016/S0193-953X(05) 70148-9. Lynch, T. R., Rosenthal, M. Z., Kosson, D. S., Cheavens, J. S., Lejuez, C. W., & Blair, R. J. (2006). Heightened sensitivity to facial expression of emotion in borderline personality disorder. Emotion, 6, 647–655. http://dx.doi.org/10.1037/1528-3542.6.4.647. Maurex, L., Lekander, M., Nilsonne, A., Andersson, E. E., Asberg, M., & Ohman, A. (2010). Social problem solving, autobiographical memory, trauma, and depression in women with borderline personality disorder and a history of suicide attempts. British Journal of Clinical Psychology, 49, 327–342. http://dx.doi.org/10.1348/014466509X454831. McCloskey, M. S., New, A. S., Siever, L. J., Goodman, M., Koenigsberg, H. W., Flory, J.D., et al. (2009). Evaluation of behavioral impulsivity and aggression tasks as endophenotypes for borderline personality disorder. Journal of Psychiatric Research, 43, 1036–1048. http://dx.doi.org/10.1016/j.jpsychires.2009.01.002. Merkl, A., Ammelburg, N., Aust, S., Roepke, S., Reinecker, H., Trahms, L., et al. (2010). Processing of visual stimuli in borderline personality disorder: A combined behavioural and magnetoencephalographic study. International Journal of Psychophysiology, 78, 257–264. http://dx.doi.org/10.1016/j.ijpsycho.2010.08.007. Mier, D., Lis, S., Esslinger, C., Sauer, C., Hagenhoff, M., Ulferts, J., et al. (2012). Neuronal correlates of social cognition in borderline personality disorder. Social Cognitive and Affective Neuroscience. http://dx.doi.org/10.1093/scan/nss028 (Advance online publication). Minzenberg, M. J., Poole, J. H., & Vinogradov, S. (2006). Social–emotion recognition in borderline personality disorder. Comprehensive Psychiatry, 47, 468–474. http://dx.doi.org/10.1016/j.comppsych.2006.03.005. New, A. S., aan het Rot, M., Ripoll, L. H., Perez-Rodriguez, M. M., Lazarus, S., Zipursky, E., et al. (2012). Empathy and alexithymia in borderline personality disorder: Clinical and laboratory measures. Journal of Personality Disorders, 26, 660–675. http://dx.doi.org/10.1521/pedi.2012.26.5.660. New, A. S., Goodman, M., Triebwasser, J., & Siever, L. J. (2008). Recent advances in the biological study of personality disorders. The Psychiatric Clinics of North America, 31, 441–461. http://dx.doi.org/10.1016/j.psc.2008.03.011. New, A. S., Hazlett, E. A., Newmark, R. E., Zhang, J., Triebwasser, J., Meyerson, D., et al. (2009). Laboratory induced aggression: A positron emission tomography study of aggressive individuals with borderline personality disorder. Biological Psychiatry, 66, 1107–1114. http://dx.doi.org/10.1016/j.biopsych.2009.07.015. Newman, L. K., Stevenson, C. S., Bergman, L. R., & Boyce, P. (2007). Borderline personality disorder, mother–infant interaction and parenting perceptions: Preliminary findings. Australian and New Zealand Journal of Psychiatry, 41, 598–605. http://dx.doi.org/10.1080/00048670701392833. Phan, K. L., Coccaro, E. F., Angstadt, M., Kreger, K. J., Mayberg, H. S., Liberzon, I., et al. (2013). Corticolimbic brain reactivity to social signals of threat before and after sertraline treatment in generalized social phobia. Biological Psychiatry, 73, 329–336. http://dx.doi.org/10.1016/j.biopsych.2012.10.0003. Platt, J., Spivak, G., & Bloom, M. (1971). Means-end problem solving procedure (MEPS): Manual and tentative norms. Philadelphia: Department of Health Sciences, Hahnemann Medical College and Hospital. Preißler, S., Dziobek, I., Ritter, K., Heekeren, H. R., & Roepke, S. (2010). Social cognition in borderline personality disorder: Evidence for disturbed recognition of the emotions, thoughts, and intentions of others. Frontiers in Behavioral Neuroscience, 4, 1–8. http://dx.doi.org/10.3389/fnbeh.2010.00182. Renneberg, B., Herm, K., Hahn, A., Staebler, K., Lammers, C. H., & Roepke, S. (2011). Perception of social participation in borderline personality disorder. Clinical Psychology and Pyschotherapy. http://dx.doi.org/10.1002/cpp.772 (Advanced online publication). Rosenthal, M. Z., Gratz, K. L., Kosson, D. S., Cheavens, J. S., Lejuez, C. W., & Lynch, T. R. (2008). Borderline personality disorder and emotional responding: A review of the research literature. Clinical Psychology Review, 28, 75–91. http://dx.doi.org/ 10.1016/j.cpr.2007.04.001. Ruocco, A.C., Medaglia, J.D., Tinker, J. R., Ayaz, H., Forman, E. M., Newman, C. F., et al. (2010). Medial prefrontal cortex hyperactivation during social exclusion in borderline personality disorder. Psychiatry Research: Neuroimaging, 181, 233–236. http://dx.doi.org/10.1016/j.pscychresns.2009.12.001.

S.A. Lazarus et al. / Clinical Psychology Review 34 (2014) 193–205 Sanislow, C. A., Grilo, C. M., Morey, L. C., Bender, D. S., Skodol, A. E., Gunderson, J. G., et al. (2002). Confirmatory factor analysis of DSM-IV criteria for borderline personality disorder: Findings from the Collaborative Longitudinal Personality Disorders Study. The American Journal of Psychiatry, 159, 284–290. http://dx.doi.org/10.1176/appi.ajp.159.2.284. Saxe, R. (2006). Uniquely human social cognition. Current Opinion in Neurobiology, 16, 235–239. http://dx.doi.org/10.1016/j.conb.2006.03.001. Saxe, R., & Kanwisher, N. (2003). People thinking about people: The role of the temporo-parietal junction in “theory of mind”. NeuroImage, 19, 1835–1842. Schaffer, Y., Barak, O., & Rassovsky, Y. (2013). Social perception in borderline personality disorder: The role of context. Journal of Personality Disorders. http://dx.doi.org/10.1521/pedi_2013_27_090 (Advance online publication). Scott, L. N., Levy, K. N., Adams, R. B., & Stevenson, M. T. (2011). Mental state decoding abilities in young adults with borderline personality disorder traits. Personality Disorders: Theory, Research, and Treatment, 2, 98–112. http://dx.doi.org/10.1037/a0020011. Segal, H. G., Westen, D., Lohr, N. E., Silk, K. R., & Cohen, R. (1992). Assessing object relations and social cognition in borderline personality disorders from stories told to the picture arrangement subtest of the WAIS-R. Journal of Personality Disorders, 6, 458–470. Shin, L. M., Wright, C. I., Cannistraro, P. A., Wedig, M. M., McMullin, K., Martis, B., et al. (2005). Archives of General Psychiatry, 62, 273–281. Sieswerda, S., Barnow, S., Verheul, R., & Arntz, A. (2013). Neither dichotomous nor split, but schema-related negative interpersonal evaluations characterize borderline patients. Journal of Personality Disorders, 27, 36–52. http://dx.doi.org/ 10.1521/pedi.2013.27.1.36. Simeon, D., Knutelska, M., Smith, L., Baker, B. R., & Hollander, E. (2007). A preliminary study of cortisol and norepinephrine reactivity to psychosocial stress in borderline personality disorder with high and low dissociation. Psychiatry Research, 149, 177–184. http://dx.doi.org/10.1016/j.psychres.2005.11.014. Skodol, A. E., Pagano, M. E., Bender, D. S., Shea, M. T., Gunderson, J. G., Yen, S., et al. (2005). Stability of functional impairment in patients with schizotypal, borderline, avoidant, or obsessive–compulsive personality disorder over two years. Psychological Medicine, 35, 443–451. http://dx.doi.org/10.1017/S003329170400354X. Soloff, P. H., & Fabio, A. (2008). Prospective predictors of suicide attempts in borderline personality disorder at one, two, and two-to-five year follow-up. Journal of Personality Disorders, 22, 123–134. http://dx.doi.org/10.1521/pedi. 2008.22.2.123. Staebler, K., Renneberg, B., Stopsack, M., Fiedler, P., Weiler, M., & Roepke, S. (2011). Facial emotional expression in reaction to social exclusion in borderline personality disorder. Psychological Medicine, 41, 1929–1938. http://dx.doi.org/10.1017/S0033291711000080. Theodoridou, A., Rowe, A.C., Penton-Voak, I. S., & Rogers, P. J. (2009). Oxytocin and social perception: Oxytocin increases facial trustworthiness and attractiveness. Hormones and Behavior, 56, 128–132. http://dx.doi.org/10.1016/j.yhbeh.2009.03.019. Tragesser, S. L., Lippman, L. H., Trull, T. J., & Barrett, K. C. (2008). Borderline personality disorder features and cognitive, emotional, and predicted behavioral reactions to teasing. Journal of Research in Personality, 42, 1512–1523. http://dx.doi.org/10.1016/j.jrp.2008.07.003.

205

Unoka, Z., Fogd, D., Füzy, M., & Csukly, G. (2011). Misreading the facial signs: Specific impairments and error patterns in recognition of facial emotions with negative valence in borderline personality disorder. Psychiatry Research, 189, 419–425. http://dx.doi.org/10.1016/j.psychres.2011.02.010. Unoka, Z., Seres, I., Aspan, N., Bodi, N., & Keri, S. (2009). Trust game reveals restricted interpersonal transactions in patients with borderline personality disorder. Journal of Personality Disorders, 23, 399–409. http://dx.doi.org/10.1521/pedi. 2009.23.4.399. Veen, G., & Arntz, A. (2000). Multidimensional dichotomous thinking characterizes borderline personality disorder. Cognitive Therapy and Research, 24, 23–45. http://dx.doi.org/10.1023/A:1005498824175. Walter, M., Bureau, J. F., Holmes, B.M., Bertha, E. A., Hollander, M., Wheelis, J., et al. (2008). Cortisol response to interpersonal stress in young adults with borderline personality disorder: A pilot study. European Psychiatry, 23, 201–204. http://dx.doi.org/10.1016/j.eurpsy.2007.12.003. Wechsler, D. (1998). Wechsler Adult Intelligence Scale (WAIS-III). (Hebrew version). London: Psychological Corporation. Wells, T. T., Clerkin, E. M., Ellis, A. J., & Beevers, C. G. (2014). Effect of antidepressant medication use on emotional information processing in major depression. American Journal of Psychiatry, 171, 195–200. http://dx.doi.org/10.1176/appi.ajp.2013.12091243. Westen, D. (1991). Social cognition and object relations. Psychological Bulletin, 109, 429–455. http://dx.doi.org/10.1037//0033-2909.109.3.429. Westen, D., Lohr, N., Silk, K., Gold, L., & Kerber, K. (1990). Object relations and social cognition in borderlines, depressives, and normals: A TAT analysis. Psychological Assessment: A Journal of Consulting and Clinical Psychology, 2, 355–364. http://dx.doi.org/10.1037//1040-3590.2.4.355. Whipple, R., & Fowler, J. C. (2011). Affect, relationship schemas, and social cognition: Self-injuring borderline personality disorder inpatients. Psychoanalytic Psychology, 28, 183–195. http://dx.doi.org/10.1037/a0023241. White, H., Flanagan, T. J., Martin, A., & Silvermann, D. (2011). Mother–infant interactions in women with borderline personality disorder, major depressive disorder, their co-occurrence, and healthy controls. Journal of Reproductive and Infant Psychology, 29, 223–235. http://dx.doi.org/10.1080/02646838.2011.576425. Williams, K. D., & Jarvis, B. (2006). Cyberball: A program for use in research on interpersonal ostracism and acceptance. Behavior Research Methods, 38, 174–180. http://dx.doi.org/10.3758/BF03192765. Wischniewski, J., & Brüne, M. (2012). How do people with borderline personality disorder respond to norm violations? Impact of personality factors on economic decision-making. Journal of Personality Disorders. http://dx.doi.org/10.1521/pedi_2012_26_036 (Advance online publication). Zanarini, M. C., Frankenburg, F. R., Dubo, E. D., Sickel, A. E., Trikha, A., Levin, A., et al. (1998). Axis I comorbidity of borderline personality disorder. American Journal of Psychiatry, 155, 1733–1739. http://dx.doi.org/10.1176/appi.ajp.161.11.2108. Zanarini, M. C., Frankenburg, F. R., Young, L. Y., Raviola, G., Reich, D. B., Hennen, J., et al. (2004). Borderline psychopathology in the first-degree relatives of borderline and Axis II comparison probands. Journal of Personality Disorders, 18, 439–447. http://dx.doi.org/10.1521/pedi.18.5.439.51327.