16
Sheila E. Crowell, Parisa R. Kaliush, and Robert D. Vlisides-Henry
Introduction
Mechanisms are processes or events that produce or catalyze change, underlie or drive an observed phenomenon, or explain why an outcome occurred (Hedström & Ylikoski, 2010; Kazdin, 2007). Not surprisingly, personality disorder and other psychopathology researchers are extremely interested in identifying mechanisms, given that much of what we study involves mental processes that are difficult to observe. For mental health practitioners, the search for mechanisms is of critical importance. Biological mechanisms associated with psychopathology may hold promise for psychiatric medication management (e.g., MacKinnon & Pies, 2006) and may be alterable with behavioral interventions (e.g., Perroud et al., 2013). Identifying psychosocial and contextual mechanisms contributing to high-risk outcomes could inform psychotherapeutic treatments or prevention efforts (Kazdin, 2007). In short, the identification of mechanistic processes that underlie change, allows for more targeted and effective therapies that can be tailored to the unique requirements of those who are in greatest need. Furthermore, assessing mechanisms during treatment could help us better understand why some therapies are effective or ineffective.
The search for mechanisms in personality disorders (PDs) is especially important. First, PDs are chronic, pervasive, costly, and a source of significant distress for those affected and their loved ones (Lieb, Zanarini, Schmahl, Linehan, & Bohus, 2004). Second, many PDs are difficult to treat and/or researchers have not conducted necessary basic research or appropriate clinical trials to establish effective intervention targets, especially for Cluster A (Bamelis, Evers, Spinhoven, & Arntz, 2014). Third, many PDs are relatively rare compared to other psychiatric diagnoses (American Psychiatric Association, 2013; Tyrer, Reed, & Crawford, 2015), which makes it difficult to conduct treatment-outcome studies without costly multi-site efforts that are challenging to fund. Finally, PDs, with the possible exception of borderline personality disorder (BPD), are among the most poorly understood psychiatric conditions, even though these diagnoses are a significant source of both morbidity and mortality (Kolla et al., 2016). Therefore, it is urgent that we advance current understanding of PDs by identifying processes that underlie the emergence and maintenance of personality pathology and drive treatment outcomes.
Despite its promise, mechanistic research is diminished by several limiting factors. As a concept, the word “mechanism” is used often and is occasionally misused in the literature. This is likely due to poor understanding of what defines a mechanism. Further, most study designs are inadequate to test mechanistic theories and the dominant analytic techniques are similarly ill-suited for establishing causal processes. In order to remedy this situation, researchers must deploy more complex study designs (e.g., longitudinal, multiple levels of analysis, random assignment) and learn more sophisticated analytic techniques (see e.g., Markon & Jonas, 2016). Finally, mechanistic research cannot occur in the absence of a theory – ideally one which has withstood many empirical challenges and the test of Occam’s razor. Thus, overly complex and/or atheoretical findings that introduce a potential mechanism must be viewed with skepticism.
In this chapter, we seek to clarify the definition of a mechanism with attention to these limiting factors and common problems as they relate to PDs. Next, we highlight dominant mechanistic theories in PD research and briefly describe methodological and analytic approaches that are well-suited to test these theories. Then, we examine empirical studies of biological, contextual, and biosocial mechanisms of risk for PDs with a critical eye. We propose that mechanistic research is critical for understanding, preventing, and treating PDs and that PD researchers should define, assess, and evaluate potential mechanisms with great care.
Terminology, Concepts, and Theories
Defining Mechanisms
The concept of a mechanism has a long history in the psychiatric literature. Freud (e.g., Freud & Breuer, 1893, p. 26) described the psychological reaction to severe trauma as “the mechanism of psychically acquired hysteria” and a causal factor in the etiology of hysterical symptoms. Thus, even early use of the term implied causality. This understanding continues to more recent definitions of a mechanism as “the cogs and wheels of the causal process through which the outcome to be explained was brought about” (Hedström & Ylikoski, 2010, p. 50). As Hedström and Ylikoski describe, there are four key elements that define a mechanism. First, a mechanism can be identified by the phenomenon or effect it produces. In other words, a mechanism includes all of the parts, operations, and their organization that, taken together, yield an observed outcome.
Second, a mechanism involves elements of a causal process that increase the probability of that observed outcome. In this regard, Hedström and Ylikoski take a more liberal definition of a mechanism than do other authors (e.g., Mahoney, 2001), who propose that a mechanism must be sufficient to produce the outcome of interest. Thus, Hedström and Ylikoski assume that mechanistic processes likely include some random elements that could produce different outcomes, particularly in the social sciences.
Third, a mechanism has structure, and “when a mechanism-based explanation opens the black box, it discloses this structure” (Hedström & Ylikoski, 2010, p. 51). Thus, a mechanistic description reveals how the outcome occurs, which includes the participating entities as well as their properties, activities, and relations. Importantly, once these components of a mechanism are revealed, it is possible to delve into a series of subsequent investigations into each of the component parts and their role in the mechanistic process (e.g., one might ask whether it is possible to prevent or change the outcome by altering certain components).
Finally, there is a hierarchy inherent in mechanisms and the scientific disciplines that seek to elucidate mechanistic processes. Consequently, what we define as a mechanism in the social and psychological sciences invariably includes elements that would be broken down further by other scientific disciplines. This progression continues down to the most elemental physical processes that can be reduced no further. Importantly, it is not necessary to elucidate every component of the causal process in order to identify and study a psychological mechanism. Even if we are unable – or neglect to delineate – each of the component entities and activities, a mechanistic account serves as a framework for understanding how an outcome came to be.
Key Concepts
Importance of Theory
The search for a mechanism requires rigorous gathering of empirical evidence in order to differentiate a verifiable mechanistic explanation from “mechanism-based storytelling” (Hedström & Ylikoski, 2010, p. 53). As a result, the process of establishing a mechanism is lengthy and begins with theory. This theory serves as one account of processes within the black box while simultaneously hypothesizing that other competing explanations are less plausible. Over time, theories are refined and mechanistic explanations reveal new black boxes that we must open and examine. Freud’s mechanistic account provides a good example. He advanced a theory that trauma is the causal mechanism of most acquired anxiety and, furthermore, through hypnosis and other techniques, it is possible to identify traumas and/or other repressed conflicts that are the source of current distress (Freud & Breuer, 1893). This prompted research into associations between trauma exposure and psychiatric symptoms, such as cases of “shell shock” during the First World War (Myers, 1915), and combat fatigue during the Second World War (Saul, 1945).
As the field advanced, scientists asked more sophisticated questions and began to test animal models of stress exposure. For example, Garattini, Giacalone, and Valzelli (1967) found that mice who were kept in isolation for four weeks became aggressive relative to those raised communally. At the end of the four-week experiment, these mice were injected with tranylcypromine (a monoamine oxidase inhibitor [MAO]) and their brains were analyzed to examine how serotonin (5HT) and the serotonin metabolite 5HIAA were processed. The authors found an increase in 5HT turnover among isolated relative to communal mice, which they implied was a potential mechanism linking isolation stress with later aggressive behavior. Psychological research on posttraumatic stress disorder (PTSD) also progressed and researchers began to ask the question of who develops PTSD following a traumatic event. Not surprisingly, meta-analytic findings revealed considerable heterogeneity among those with a PTSD diagnosis, their pretrauma histories, and trauma-concurrent stressors/supports (Brewin, Andrews, & Valentine, 2000). Current research on mechanisms linking trauma exposure to PTSD is focused on a number of biological and social risk factors, including genes, epigenetic regulation, inflammation, psychophysiology, neurocircuitry, prior trauma or childhood family adversity, preexisting mental disorders, and lack of social support (see e.g., Admon, Milad, & Hendler, 2013; Bromet, Atwoli, Kawakami, & Navarro-Mateu, 2017; McLaughlin & Lambert, 2017; Shalev, Liberzon, & Marmar, 2017). Thus, the black box of how trauma leads to psychopathology, and for whom, is becoming more transparent and nuanced.
Design and Analytic Considerations
A mechanistic theory provides a causal hypothesis of how several elements, processes, and activities operate together to produce – or increase the probability of – an outcome. However, it is a challenge to test causal theories in PDs and other complex psychological disorders. Thankfully, modern research has transcended early theories involving simple etiological causes, such as a single gene for psychopathy (although psychopathy is highly heritable; Larsson, Andershed, & Lichtenstein, 2006; Viding, Blair, Moffitt, & Plomin, 2005) or trauma as the root of BPD (although trauma is a known risk factor for BPD; Lieb et al., 2004). Instead of searching for the cause of PDs, researchers now seek to understand pieces of a complex causal puzzle in which many elements contribute, probabilistically, to the end result. This requires drawing careful conclusions from well-designed experiments.
Historically, researchers have sought to test causal theories using statistical methods (Kazdin, 2007). However, statistics alone are insufficient for identifying causal processes unless certain design considerations are met, such as random assignment to experimental or control conditions, careful control of confounding variables, and manipulation of the independent variable (see e.g., Crowell et al., 2017). In PD research, this is most often accomplished within clinical trials. For example, in one clinical trial of treatment for BPD (Clarkin, Levy, Lenzenweger, & Kernberg, 2007), participants were randomized to three different interventions that were similar in the amount of contact provided, quality of providers, and several other variables. The results revealed that targeting specific clinical problems in therapy led to changes in key outcomes for some therapies over others. Transference focused therapy, for example, was uniquely effective at reducing irritability and verbal and direct assault. The authors hypothesized that repeatedly focusing on self-control in the context of the therapist–patient relationship may have been a mechanism by which this change occurred.
Similarly, Gratz and colleagues (Gratz, Bardeen, Levy, Dixon-Gordon, & Tull, 2015) examined changes in emotion dysregulation as a mechanism of change in a group therapy targeting emotion regulation deficits. They found that reductions in emotion dysregulation mediated the effects of group therapy on BPD symptoms, such as self-inflicted injury. Clinical trials offer a powerful opportunity to test mechanistic theories because they allow random assignment to an enhanced experience (given that random assignment to stressful environments is unethical). However, this design often involves many synergistic elements and processes – individual therapy, group therapy, medication management, therapist–client match, client preferences for treatment style – making it difficult to disentangle key components contributing to change (see e.g., Ahn & Wampold, 2001). Other common designs include random assignment to different tasks or examining within-person changes to laboratory stressors such as conflict or rejection (e.g., Crowell et al., 2005, 2017). These designs allow for a more microanalytic approach to testing mechanistic theories. A limitation of these approaches is that they only allow for tests of proximal causes within a larger causal sequence.
A wide range of analytic and statistical approaches have been developed and used to test causal theories. The most common include mediation analysis, structural equation modeling, and other multivariate approaches (e.g., Joreskog, Sorbom, & Magidson, 1979; Lowry & Gaskin, 2014; Preacher & Hayes, 2004). More recently, researchers have introduced dynamic causal modeling – a Bayesian approach – and convergent cross mapping – a dynamic systems technique designed to detect causal relationships within time-series data (Clark et al., 2015; Stephan et al., 2010). However, we must evaluate the causal promises of these methods with a critical eye, remembering that answers will only be meaningful in the context of a well-reasoned question and study design. Indeed, in one of the most widely cited articles on causal modeling, Bentler (1980, p. 420) states that the word “cause” is not intended to convey any philosophical meaning beyond “a hypothesized unobserved process” and that other terms, such as process or system modeling would work equally well. Thus, just like the theories they are designed to test, statistical methods that underlie mechanism-based research are both fraught and full of promise.
Mechanistic Theories and Personality Disorders
Clearly, there are many challenges inherent to defining, studying, and testing mechanisms. If we accept the broadest definitions of the term, a mechanism can be examined at almost any level of analysis and can include almost any combination of elements within a causal sequence of events. This opens up a veritable gold mine of research questions for scientists interested in understanding how PDs emerge, are maintained, and change over time or with treatment. The risk, however, is that we have identified a term that means everything and, as a result, is meaningless. Thus, if we seek to advance PD research, we must ground our mechanistic work in testable (and falsifiable) theories.
There are several theories currently at the forefront of PD research. A majority of these are based on dimensional, trait-based conceptualizations of personality and PDs. Dimensional theorists seek to explain psychopathology in terms of broad, transdiagnostic traits, such as internalizing/externalizing and the Five-Factor Model (Kotov et al., 2017; Krueger & Markon, 2014; Wright et al., 2012). This perspective holds great promise for mechanistic research, since underlying mechanisms of risk likely cut across diagnostic categories (Beauchaine & McNulty, 2013; Kotov et al., 2017). Indeed, as Kupfer, First, and Regier (2002, p. xviii) noted, “not one laboratory marker has been found to be specific in identifying any of the DSM-defined syndromes. Epidemiologic and clinical studies have shown extremely high rates of comorbidities among the disorders, undermining the hypothesis that the syndromes represent distinct etiologies.” This observation, along with research over the past 15 years, suggests that common traits underlie many DSM diagnoses, including PDs (Caspi et al., 2014; Cuthbert & Insel, 2013; Kotov et al., 2017). For PD theorists, however, connections between general personality structure and diagnosable PDs are especially clear because PDs appear to represent an extreme variant of normative personality dimensions (Crowell & Kaufman, 2016; Miller, Lyman, Widiger, & Leukefeld, 2001; Widiger & Simonsen, 2005).
In addition to dimensional approaches, PD researchers have increasingly focused on etiology and developmental precursors to PDs (Bornovalova, Lejuez, Daughters, Rosenthal, & Lynch, 2005; Crowell, Beauchaine, & Linehan, 2009; De Fruyt & De Clercq, 2014). This work also emerges from a trait-based conceptualization of psychopathology with a specific emphasis on early biologically-based temperament, parent–child relationships (e.g., attachment, interaction patterns), and biology × environment interactions. As with dimensional approaches, a major focus of this work is on emotional processes, such as emotion dysregulation, and mechanisms that underlie its development, including research on emotional instability, poor emotional awareness, mood-dependent impulsive behavior, emotional lability, and other forms of dysregulated emotions and behavior. Thus, although there are a range of potential mechanisms that are relevant to PD research, emotional processes are a focus of this review.
Biological Mechanisms
Researchers use a number of techniques to test biological mechanisms of PDs and PD development, including neuroimaging, psychophysiological, neurotransmitter activity, genetic, and epigenetic processes. Understandably, each of these methods can only elucidate a few components of the many complex mechanisms that underlie PDs. For example, neuroimaging techniques are useful for revealing biological responses that occur during emotion dysregulation (Doll et al., 2013). Even though imaging and many other biological techniques can only reveal part of a mechanistic process, biomarkers serve as potential targets in treatment research and help scientists understand emotional and psychological processes at another level of analysis.
Neuroimaging
Researchers have used functional magnetic resonance imaging (fMRI), MRI, and electroencephalography (EEG) to examine emotional processes in the moment. Specifically, those with PDs tend to have distinct activation of brain areas related to emotional processing. For instance, Doll and colleagues (2013) used fMRI to examine connectivity between the default mode network, salience network, and central executive network. They hypothesize that connections between these brain networks form a foundation for emotion regulation. Specifically, these areas are activated by emotions, cognitions, and behaviors – thus, connectedness allows for appropriate coordination of neural activity. The researchers found that those with BPD (compared to controls) had abnormal connectivity between the networks, suggesting a possible mechanism for emotion dysregulation. Other researchers have shown that connectivity between neural regions is important for emotion regulation. For those with schizotypal PD, there appears to be altered frontotemporal activity and connectivity between these areas and other neural regions (Fervaha & Remington, 2013).
In addition to neural pathways, activity in specific regions may also be associated with emotion dysregulation. The amygdala is an important region for emotions (particularly fear) and their regulation, ultimately making it a candidate mechanism of PDs. A smaller amygdala with reduced functioning has been associated with psychopathy (Moul, Killcross, & Dadds, 2012). From this, one may tentatively conclude that limited emotion regulation might underlie risk for psychopathy. Researchers have also found that the anterior cingulate cortex (ACC) and amygdala of those with BPD displayed heightened activity (compared to controls), both at rest and in response to fearful faces (Mitchell, Dickens, & Picchioni, 2014). Relatedly, MacKinnon and Pies (2006) reviewed structural MRI techniques and found that women with BPD had reduced hippocampal volume and elevated blood oxygen levels in the amygdala compared to controls. In a review, Susman (2006) discussed how abnormal amygdala functioning might mediate the relation between early-life trauma and emotion dysregulation. Finally, Hajcak, MacNamara, and Olvet (2010) have used EEG to determine that the specific event-related potentials (ERP) P300 and the late positive potential (LPP) are linked to emotion regulation. Future research should examine how these brain regions might be causally linked to emotion dysregulation and PDs.
Mixed neurological findings also have been found in children and adolescents, though the literature is still emerging. Goodman, Mascitelli, and Triebwasser (2013) compared the neurobiology literature on adult-onset and adolescent-onset BPD. However, they found minimal data on brain abnormalities in adolescent-onset BPD compared to controls. That is, there were no clear differences in ACC, amygdala, and hippocampus size and functionality or ERP P300. In another review, Brunner, Henze, Richter, and Kaess (2015) also did not find any structural or functional differences but they do report clear differences in limbic system gray matter volume and functionality for children with BPD, suggesting differences in emotion regulation from an early age. These limbic gray matter changes could be due to an interaction with early-life stress and abuse (Ensink, Biberdzic, Normandin, & Clarkin, 2015). Other reviews have found similarly mixed results in regards to brain structures, EEG findings, and gray matter differences (Winsper et al., 2016), though some older articles still need to be replicated (see e.g., Deckel, Hesselbrock, & Bauer, 1996). Thus, there is a clear need for further research on neurobiological mechanisms of youth PDs.
These studies all support the hypothesis that diminished amygdala size and functioning may play a mechanistic role in PDs, though this may be limited to adults. However, due to study design limitations, sample sizes, and lack of replication of key findings, we cannot be certain that we have identified key components of the mechanistic pathway (see also van Zutphen, Siep, Jacob, Goebel, & Arntz, 2015 for a critical review of imaging findings in BPD). Nevertheless, these findings bring us closer to a better understanding of emotional processes in PDs.
Neurochemistry
Researchers have found that neurochemical and hormonal abnormalities are associated with emotion dysregulation. Although a thorough review is beyond the scope of this chapter, neurotransmitter and neuroendocrine dysfunction are a major focus of PD research (for a review, see e.g., Bridgett, Burt, Edwards, & Deater-Deckard, 2015). Briefly, monoamine oxidase-A and -B (MAO-A, MAO-B) are a group of enzymes that catalyze monoamines. Low levels of MAO-B have been found in antisocial personality disorder (ASPD) and BPD (Zuckerman & Kuhlman, 2000). In a study by Kolla and colleagues (2016), researchers found that MAO-A levels in the prefrontal cortex and ACC were elevated among those with BPD compared to control individuals.
There is also an extensive literature linking neurotransmitter function, psychopathology, and PDs (for reviews, see e.g., Kenna et al., 2012; Martin, Ressler, Binder, & Nemeroff, 2010). Researchers have found that serotonin is critical for regulating emotions, particularly aggression, with reduced serotonin levels predicting greater frequency of antisocial behaviors (Trull, Stepp, & Durrett, 2003). Lee (2006) found that serotonin and GABA interact with one another to affect amygdala activation and impair self-regulation. Susman (2006) found that attenuation of the serotonergic system, abnormalities in the gamma aminobutyric (GABA) system, and reduced cortisol levels were all linked to antisocial behaviors and ASPD. Finally, alterations in vasopressin and oxytocin may mediate the established link between early-life trauma and PD development (Heinrichs, von Dawans, & Domes, 2009). Thus, neurotransmitters appear to play a central role in self-regulation, psychopathology, and PDs (see also, Strauman, 2017).
Additionally, HPA axis activation (i.e., corticotropin releasing factor [CRF] and cortisol) is related to emotion dysregulation and has been a focus of research on depression and anxiety (Pagliaccio et al., 2015; Stetler & Miller, 2011). Lee (2006) found that CRF, serotonin, and GABA levels interact with one another to affect amygdala activity, ultimately hampering regulatory capacity. In one review, Hostinar, Sullivan, and Gunnar (2014) found a literature consensus that cortisol levels can be regulated through social support, though this process was ultimately moderated by levels of oxytocin, vasopressin, and sympathetic neurotransmitters (e.g., norepinephrine, epinephrine). In one interesting animal study, Butler and colleagues (Butler, Ariwodola, & Weiner, 2014) found that socially isolated rats showed disrupted HPA axis function, anxiety-like behaviors, and were more likely to develop a preference for and over-use of EtOH (alcohol). Thus, HPA-axis dysfunction is one potential link in the causal chain from social isolation to alcohol abuse and other forms of psychopathology, which has clear relevance for PDs (Butler, Karkhanis, Jones, & Weiner, 2016; Crowell, 2016; Mushtaq, Shoib, Shah, & Mushtaq, 2014).
While some have found no differences in cortisol activity for adolescents with BPD (Winsper et al., 2016), other reviews have shown attenuated cortisol reactivity to stress in adolescents (Brunner et al., 2015; Ensink et al., 2015; Goodman et al., 2013), again evidencing the inconsistency in the biological findings for PD youths. Few other neurochemical differences have been found, as most authors report either a lack of clear differences between PD youth and control youths or simply a dearth of literature (e.g., Brunner et al., 2015).
Other Physiology
A variety of other physiological processes and markers have been linked to regulatory capacity and PDs. For example, those with BPD tend to have lower cholesterol and leptin levels compared with controls (Trull et al., 2003). In a comprehensive review, Thayer and Sternberg (2006) found that reduced vagal tone (i.e., low heart rate variability [HRV]/respiratory sinus arrhythmia [RSA]), an established marker of regulatory capacity, is linked to greater levels of inflammatory markers (e.g., interleukin-6) and cortisol and reduced levels of glucose. This review suggests that unhealthy physiology (e.g., inflammation, cortisol) is associated with emotion dysregulation, which in turn is associated with psychopathology and PDs. Further, in a meta-analysis, Koenig, Kemp, Feeling, Thayer, and Kaess (2016) compared resting state HRV in BPD individuals compared to controls. The results showed a dosage effect of BPD symptoms on HRV, with lower HRV being related to more BPD traits. The autonomic nervous system is also involved in antisocial symptoms. Those with ASPD and incarcerated individuals tend to have reduced autonomic arousal (i.e., less heart rate reactivity) compared to controls (Susman, 2006). In the same review, the author found that children who display risk for ASPD have reduced HRV when challenged compared to controls, suggesting a lack of regulatory capacity. In sum, a variety of physiological markers are related to emotion dysregulation and PDs.
There is an extensive literature on youth psychopathology and psychophysiological indices of risk (see Beauchaine, 2001). However, few researchers have attempted to link these findings to theories of PD development. In one study, Raine and colleagues found that low resting heart rate at age 3 was a significant predictor of aggression and antisocial behaviors at age 11 (Raine, Venables, & Mednick, 1997). In our own work, we have also examined how peripheral physiology is associated with PD risk. For instance, we found that self-injuring adolescents scored higher on measures of emotion dysregulation, externalizing psychopathology, and BPD symptoms, and also had attenuated electrodermal responding (EDR) compared to depressed adolescents (Crowell et al., 2012). This suggests that EDR may be one mechanism of risk for impulsivity and externalizing traits among girls at risk for BPD, although further research is needed. When examining dyads consisting of depressed adolescents and their mothers, we found that depressed and self-injuring adolescents showed moment-to-moment withdrawal in RSA in response to aversive maternal behaviors and their mothers showed a similar pattern in response to adolescent aversive behaviors. In contrast, control adolescents and their mothers showed RSA increases in response to aversive behaviors, which possibly reflects better emotion regulation in the face of interpersonal stress (Crowell et al., 2014). Although many participants in these studies had only subthreshold PD traits, research with high-risk adolescents is important for bridging the gap between early vulnerability factors and a later PD diagnosis.
Genetics
Researchers have examined a variety of candidate genes and genetic factors that underlie emotion dysregulation and PDs. Many of these genes are related to the production, transport, and degradation of neurotransmitters (e.g., MAO-A, serotonin), but sex chromosomes also appear to play a role. For example, the male sex chromosome puts males at greater risk for aggressive behaviors (Eme, 2007). Those with a variation of the MAO-A production gene such that there is reduced MAO-A production tend to be at greater risk for aggression, conduct disorder, and ASPD (Lee, 2006; Susman, 2006). Researchers have also found that BPD, depression, and emotional lability co-aggregate in families (MacKinnon & Pies, 2006), potentially suggesting a common genetic pathway. Canli, Ferra, and Duman (2009) showed that genetic variations resulting in less production of 5-HTTLPR, COMT, and MAO-A are linked to top-down cortical emotion modulation – that is, prefrontal involvement in emotion regulation. The effects of genes may be differentially impactful throughout the lifespan. For instance, Bornovalova, Hicks, Iacono, and McGue (2009) found that the heritability of BPD is slightly higher for those aged 14–24 than for older adults. Genes have been linked clearly to temperaments that might predispose the youth to PD risk (Brunner et al., 2015; Ensink et al., 2015). Additionally, a few candidate genes may put children and adolescents at risk for BPD symptoms, including the short 5-HTTLPR (similar to adults), the oxytocin receptor, and FKBP5 (Winsper et al., 2016). In a recent study, Bornovalova and colleagues (2018) found that genetic differences primarily accounted for the comorbidity between BPD and substance use disorders, while BPD comorbidity with other disorders was largely due to environmental influences. This study furthers understanding of the common genetic (or epigenetic) mechanisms to psychiatric comorbidity in PDs.
Epigenetics
Researchers studying epigenetic processes typically examine genetic methylation, the process by which gene function is more or less activated, although specific effects vary greatly by gene, situation, and amount of methylation (Bird, 2002). Perhaps unsurprisingly, methylation of the MAO-A and MAO-B genes appears to be related to emotion dysregulation and PD risk. Dammann and colleagues (2011) showed that MAO-A and MAO-B methylation predicted BPD risk for females only. Furthermore, hypermethylation of the MAO-A promoter may lead to downregulation of MAO-A, ultimately predicting reduced serotonin concentration and ASPD risk (Checknita et al., 2015). The methylation of other genes has also been linked to emotion dysregulation: S-COMT methylation has predicted BPD (Dammann et al., 2011) and methylation of the oxytocin receptor gene has predicted callousness and lack of sociality (Kumsta, Hummel, Chen, & Heinrichs, 2013).
Contextual and Environmental Mechanisms
There are countless contextual and environmental risk factors, ranging from childhood abuse (see e.g., Belsky et al., 2012) to low socioeconomic status (see e.g., Cohen et al., 2008), that are associated with development and maintenance of PDs. Researchers sometimes describe these as mechanisms; however, not all risk factors meet the definition of a mechanism. Potential contextual and environmental mechanisms must be linked to theories of change and/or processes that underlie the emergence of PDs. Research exploring contextual and environmental mechanisms of risk are crucial for informing psychosocial prevention and intervention (Crowell et al., 2013). However, research to date is limited by difficulties determining what constitutes a mechanism in the social sciences and challenges with determining causality among associated variables, especially if they are assessed concurrently. Also, some authors use contextual and environmental mechanisms interchangeably, which can confuse interpretation of findings and limit our ability to apply findings to prevention and intervention efforts. To limit this confusion, we describe contextual and environmental mechanisms separately. We define contextual mechanisms as processes in an individual’s local, daily environment that directly influence their health and well-being. These processes are malleable and shift frequently (e.g., parenting behaviors). In contrast, environmental mechanisms are more stable and encompass broader societal factors, such as neighborhood violence and socioeconomic status (SES).
Contextual Mechanisms
Early Maternal Withdrawal
There are an increasing number of prospective studies examining the influences of early caregiving interactions on the development of psychopathology. Primarily, this research focuses on early mother–infant interaction patterns, and addresses such constructs as early separation, disrupted communication, and disorganized attachment. Maternal withdrawal to infant attachment cues emerges repeatedly as a strong predictor of BPD and conduct disorder symptoms, as well as suicidality and self-injury in adolescence (Lyons-Ruth, 2008; Lyons-Ruth, Bureau, Easterbrooks, Obsuth, & Hennighausen, 2013; Lyons-Ruth, Bureau, Holmes, Easterbrooks, & Brooks, 2013; Steele & Siever, 2010; Stepp, Lazarus, & Byrd, 2015). Maternal withdrawal is characterized by a general lack of interaction with an infant, including a lack of greeting and comforting, delayed or cursory responding, redirecting the infant’s attention from the mother to toys, and engaging primarily in distanced interactions (e.g., interacting from across the room; Lyons-Ruth, Bureau, Easterbrooks, et al., 2013).
Interestingly, maternal withdrawal to infant attachment cues has surfaced as a stronger predictor of adolescent borderline and conduct disorder symptoms than has maternal negative-intrusive behavior, despite consistent links between negative-intrusive behavior and abuse, and abuse and borderline pathology (Lyons-Ruth, 2008). For instance, Lyons-Ruth, Bureau, Holmes, et al. (2013) found that maternal withdrawal during an infant attachment assessment at 18 months accounted for the relation between clinician referral due to concerning quality of care and offspring borderline and conduct disorder features around age 18 years. The effect of maternal withdrawal on later offspring borderline and conduct disorder features was independent of, and additive to, the severity of childhood abuse (Lyons-Ruth, Bureau, Holmes, et al., 2013). Thus, child abuse is an important risk factor in the development of personality disorders, but it is more likely that abusive experiences interact with early maternal interactions (i.e., environment × environment interaction), rather than acting alone, to confer risk for pathology (Caspi et al., 2002; Fruzzetti, Shenk, & Hoffman, 2005; Neuhaus & Beauchaine, 2017).
Extended maternal separation has also been identified as a potential mechanism in the development of PDs (Chanen & Kaess, 2012; Crawford, Cohen, Chen, Anglin, & Ehrensaft, 2009; Steele & Siever, 2010). Specifically, extended maternal separation (i.e., more than one month) from the infant during the first five years was found to be a significant predictor of offspring borderline personality pathology among a large random community sample (Crawford et al., 2009). Not only did extended early maternal separation predict the presence of adolescent borderline symptoms, it predicted significantly slower declines of these symptoms as the offspring progressed through normative maturation and socialization processes (Crawford et al., 2009). Interestingly, separation due to divorce or death was not predictive of offspring borderline symptoms (Crawford et al., 2009). Thus, it is difficult to determine if the separation itself catalyzed the development of personality pathology. Because separation due to other reasons, such as mothers leaving for personal reasons or infants being sent away for extended stays with relatives, was predictive of borderline symptoms, it is more likely that extended separation is a risk factor when it is due to maternal withdrawal (i.e., lack of maternal investment in caregiving).
Invalidating Interaction Patterns
Invalidating parent–child interactions are characterized by parental rejection or minimization of children’s emotional expressions, especially those that are overwhelming for the family to manage (Crowell, Yaptangco, & Turner, 2016; Linehan, 1993). This intolerance toward a child’s emotional expression communicates to the child that their experiences are unreasonable, and that they must cope independently with their distress (Linehan, 1993). Consequently, the child does not learn basic emotion regulation skills, and instead uses increasingly labile interaction patterns to garner support and validation from caregivers (Crowell et al., 2009). This feedback loop of invalidation and extreme emotional lability may recur over many years, disrupting family relationships and increasing the child’s risk for PD development (Crowell et al., 2009, 2013, 2016; Linehan, 1993).
Hallquist, Hipwell, and Stepp (2015) prospectively investigated invalidating parenting, poor self-control (e.g., inability to control temper during arguments), and negative emotionality among girls aged 5–14 years as predictors of BPD at 14–17 years of age. These researchers found that all three factors were predictive of borderline personality symptoms at age 14 (Hallquist et al., 2015). Importantly, results revealed a reciprocal effect of poor self-control and invalidating parenting on each other in their prediction of borderline personality symptoms at age 14 (Hallquist et al., 2015). These findings highlight the importance of studying bidirectional parent–child influences on the development of personality pathology, and support the theory that invalidating interaction patterns are prominent mechanisms in the development of emotion dysregulation and PDs.
Coercive Interaction Patterns
Coercion theory (Patterson, 1982) was developed separately from Linehan’s (1993) invalidating environment theory, but also involves reinforcement of extreme negative affectivity and emotion dysregulation (Beauchaine, Klein, Crowell, Derbidge, & Gatzke-Kopp, 2009; Crowell et al., 2016; Dishion, Duncan, Eddy, Fagot, & Fetrow, 1994; Snyder, Schrepferman, & St. Peter, 1997). During coercive interactions, parents of emotionally aroused children match or exceed the child’s aversiveness, who, in turn, matches or exceeds the arousal of their parent (Beauchaine et al., 2009). This escalation continues until the antagonistic interaction terminates, negatively reinforcing aggression, emotional lability, and autonomic arousal (Beauchaine et al., 2009). Individuals who are raised within coercive environments may not acquire the basic skills of discriminating and labeling emotions or managing strong emotions, which increases risk for PDs (Fruzzetti et al., 2005).
Initially, coercion theory was developed and tested among adolescent males who displayed externalizing behavior problems and were at risk for developing antisocial PD (see e.g., Patterson, DeBaryshe, & Ramsey, 1990). However, we have examined coercive mother–child interaction patterns among self-injuring adolescents at risk for developing BPD (Crowell et al., 2013). Each mother–child dyad engaged in a ten-minute discussion about a topic of conflict, and trained research assistants coded the interactions for parental invalidation and conflict escalation as well as aversive utterances. We found that mothers of self-injuring adolescents primarily matched or escalated conflict, and would de-escalate conversations only after extreme adolescent behavior. These coercive responses contrasted with control mothers who primarily matched at the lowest level of adolescent aversiveness and de-escalated more intense utterances, regardless of adolescent behavior. Thus, we found emerging evidence of coercive parent–child interactions among emotionally dysregulated and self-injuring adolescents at risk for PD development.
Invalidation and Coercion among Adults
Adults with PD diagnoses, especially BPD, often have distressing interpersonal histories, and demonstrate continued difficulty in forming healthy relationships (see Crowell, 2016 for a review). Thus, there is an urgent need to use more sophisticated research designs and statistical analyses to investigate invalidating and coercive interaction patterns among adults with PDs. Much of research on adults with PDs focuses on individual-level factors, neglecting potential dynamic contextual influences, such as invalidating or coercive interaction patterns with romantic partners, which may serve as mechanisms underlying maintenance of PD symptoms (Chen et al., 2004; Crowell, 2016). There is growing consensus in the field of PD research that PDs demonstrate more heterotypic continuity than stability across development (Sharp & Romero, 2007; Skodol et al., 2002), which highlights the need for more longitudinal research on contextual mechanisms of PDs in adulthood (Conway, Hammen, & Brennan, 2015).
Environmental Mechanisms
Neighborhood Effects
Traditionally, neighborhood environments have been considered indirect influences in developmental pathways to personality pathology, and therefore would not be deemed mechanisms. Indeed, there are countless interwoven factors (e.g., socioeconomic status, violence exposure, pollution) associated with neighborhoods, making it nearly impossible to detect a causal relation between one factor and PD development (Caspi, Taylor, Moffitt, & Plomin, 2000). Caspi et al. (2000) conducted a nationwide study of 2-year-old twins, and found that children in deprived neighborhoods (i.e., those characterized by low car availability, greater number of single parents, and high unemployment rates) were at significantly greater risk for demonstrating emotional and behavior problems than were those in higher SES neighborhoods. Further, these neighborhood effects accounted for variability in behavior and emotional problems above and beyond genetic liability (Caspi et al., 2000). Research demonstrates consistently that children who are impulsive and raised in high-risk neighborhoods are more likely to engage in antisocial behaviors (Lynam et al., 2000; Neuhaus & Beauchaine, 2017). Thus, there is clearly an effect of neighborhood on the development of emotion dysregulation and possibly PDs. However, the exact mechanisms that catalyze development of personality pathology warrant further investigation.
Peer Group Affiliation
Peer group affiliation is often related to an individual’s neighborhood and has been shown to influence the development of personality pathology (Beauchaine et al., 2009; Dishion, McCord, & Poulin, 1999; Ingoldsby & Shaw, 2002; Nelson & Dishion, 2004; Piehler & Dishion, 2008). Certain neighborhoods may foster harmful peer group affiliations that subsequently predict delinquency, substance use, violence, and adult maladjustment (Beauchaine et al., 2009). Dishion and colleagues explored these associations and found that rejection and isolation from peers in grade school was predictive of adult antisocial behaviors, even after controlling for early academic performance and presence of antisocial behaviors (Nelson & Dishion, 2004; Piehler & Dishion, 2008). Also, Dishion et al. (1999) found that peer group interventions can inadvertently become iatrogenic when they foster “deviancy training” among high-risk adolescents. Snyder et al. (2008) found that, among boys and girls, peer “deviancy training” in kindergarten predicted conduct disorder symptoms by third grade, and acted independently of several factors, including peer coercion, child impulsivity, and child verbal ability. These contagion effects from deviant peer affiliation have been observed among self-injuring and emotionally dysregulated youth at risk for BPD as well (Beauchaine et al., 2009; Putnam & Silk, 2005). Research on peer group affiliation offers a promising avenue for psychosocial prevention and intervention of PDs, and highlights the importance of examining several related constructs when delineating environmental mechanisms in PD development.
Biology–Environment Interactions
Research examining both biological vulnerabilities and contextual/environmental risk factors has revealed a synergistic rather than additive effect on PD development (Beauchaine et al., 2009). In fact, biology × environment interactions may be present even in the absence of significant main effects (Beauchaine et al., 2009). Advanced research methods foster the examination of complex interaction models and have promoted the emergence of the biosocial model of PDs. Marsha Linehan (1993) outlined a leading biosocial theory of BPD. She hypothesized that BPD is a disorder of emotion dysregulation, which emerges when a biologically vulnerable child is faced with specific environmental risks (see also Crowell et al., 2009). As personality pathology seems to demonstrate heterotypic continuity from childhood to adulthood (Sharp & Romero, 2007; Skodol et al., 2002), it is crucial that researchers investigate biology × environment interactions across development.
Prenatal and Infancy
Prenatal adversity, epigenetic programming, and infant temperament are important processes in the onset and development of psychopathology, including PDs (Gartstein & Skinner, 2018). Epigenetics involves alterations in gene function, without changing gene structure, that occur in response to environmental exposures (Bird, 1986) – it is the molecular mechanism driving environmental influences on phenotypic outcomes associated with physical and mental health (Gartstein & Skinner, 2018). One of the most commonly studied forms of epigenetics is DNA methylation – that is, the addition of a small methyl group to a cytosine nucleotide-phosphate-guanine nucleotide (CpG) sequence which has the capacity to activate or deactivate genes (Gartstein & Skinner, 2018). Prenatal exposure to maternal stress and depression predicts greater NR3C1 (glucocorticoid receptor) gene methylation and stress, resulting in heightened HPA (cortisol) responsivity, in infants (Monk, Spicer, & Champagne, 2012; Oberlander et al., 2008). These findings point to NR3C1 methylation as a mechanism by which early events affect later-life stress and emotional lability, which is consistent with a stress and emotion dysregulation pathway to PD risk. Epigenetic mechanisms also mediate environmental influences on infant temperament (e.g., negative emotionality, extraversion, effortful control; Gartstein & Skinner, 2018). Temperament is highly heritable and directly associated with internalizing and externalizing behaviors that become more salient in infancy (Beauchaine, 2015; Bornovalova et al., 2013). Interestingly, researchers are finding that infant temperament may serve as a genetic mediator, or mechanism, driving the relation between abuse experiences and PD development (Bornovalova et al., 2013).
Childhood and Adolescence
In addition to examining infant temperament as a mediator in the relation between early abuse experiences and PD development, researchers have found that children genetically predisposed to lower production of serotonin and COMT tend to be at greater risk for harsher discipline by parents (Bridgett et al., 2015). They described these findings in both interaction and evocative terms – children naturally producing less serotonin and COMT tend to have poorer emotion regulation, evoking harsher responses from parents. This biological vulnerability then interacts with parental stress levels to result in harsher disciplining. Trait impulsivity, which is associated with an early temperamental trait of behavioral disinhibition (Bornovalova et al., 2013), is not inherently pathological or a sufficient predictor of PD development (Bornovalova, Gratz, Delaney-Brumsey, Paulson, & Lejuez, 2006; Sharma, Markon, & Clark, 2014), but may confer risk for eventual borderline and antisocial PD symptoms when interacting with environmental risk factors. For example, Lynam et al. (2000) found that impulsive 13-year-old boys were at greater risk for juvenile offending at age 17 in poor versus high SES neighborhoods, indicating that impulsivity was associated with antisocial behaviors only when paired with poor neighborhoods (i.e., those defined by census-SES data as high poverty).
Young Adulthood and Adulthood
Research on the biology × environment interactions conferring risk for PD development and maintenance into adulthood lacks cohesion with work done on infant temperament and adolescent PD development. In general, longitudinal studies on biological vulnerabilities and environmental conditions contributing to adult PDs are sparse (Conway et al., 2015), and authors rarely outline viable mechanisms, as they mainly focus on individual-level factors (Crowell, 2016). Nonetheless, there is some research exploring biosocial mechanisms in PD development among young adults and adults. For example, Stepp, Scott, Jones, Whalen, and Hipwell (2016) assessed negative affectivity (i.e., a temperamental trait) among at-risk girls across three years (ages 16–18) and found a significant interaction between negative affectivity and family adversity to predict BPD symptoms; specifically, exposure to adversity strengthened the relation between negative affectivity and BPD symptoms. Also, Perroud et al. (2011) retrospectively assessed childhood maltreatment and sexual abuse experiences among adults and found that NR3C1 methylation interacted with not only prenatal risk, but childhood abuse to predict greater HPA axis activity and, ultimately, BPD risk. Finally, in one novel study, Knoblich and colleagues (2018) examined epigenetic changes among women with BPD following treatment with Dialectical Behavior Therapy (DBT) and found an interesting epigenome × treatment interaction: patients with higher methylation of APBA3 and MCF2 responded better to DBT treatment. Although preliminary, this study introduces a unique approach to epigenetic research in adults with PDs.
Limitations and Future Directions
There are many limitations of the current literature. First, most studies focus on borderline and antisocial PDs, with relatively fewer examining Cluster A and C diagnoses. Second, there is an overwhelming focus on emotional processes, slightly less research on mechanisms underlying risky or impulsive behavior (e.g., Lawrence, Allen, & Chanen, 2010), or cognitive and attentional mechanisms (e.g., Posner et al., 2002), and even less research on mechanisms of identity-related or interpersonal distress (however, see Kaufman & Crowell, 2018; Lejuez et al., 2003). Third, there is little continuity between child, adolescent, and adult studies and few lifespan developmental theories (e.g., Hughes, Crowell, Uyeji, & Coan, 2012). This limits our understanding of early precursors for adult personality pathology or adult outcomes of child PD risk. Fourth, as with other psychopathology research, there is ongoing need for replications of mechanistic findings, systematic reviews, and meta-analyses. Finally, there are too few studies examining mechanisms of treatment response in transdiagnostic samples. As the field moves toward a more dimensional understanding of psychopathology, it may be possible to examine key PD features across diagnostic groups, including but not limited to avoidance, emotion dysregulation, impulsivity, scrupulosity, withdrawal, and suspiciousness. Thus, our understanding of PDs will only grow along with the broader focus on dimensions of risk. Future research should continue to address these limitations in order to expand and refine our mechanistic understanding of PDs.
In conclusion, mechanism-based work holds great promise for refining theories of PD etiology, inspiring new directions in research, and improving treatment. However, it is important for the field to be thoughtful in how we identify and test mechanisms. Many studies rely on simple statistical tests (e.g., mediation with cross-sectional data) or propose novel biomarkers that have not yet been studied rigorously (e.g., candidate gene studies). Unfortunately, some of these studies advance potential PD mechanisms in the absence of theory. This makes it difficult to place empirical findings within a broader theoretical framework and replicate key components of the theory. In order to improve and expand mechanistic research, it is important to take stock of current findings and delve into each new black box – carefully identifying as many component parts as possible. Given the significant suffering associated with PDs, we must leave no stone unturned in search of factors that underlie PD risk or drive treatment outcomes.
References
Admon, R., Milad, M. R., & Hendler, T. (2013). A causal model of post-traumatic stress disorder: Disentangling predisposed from acquired neural abnormalities. Trends in Cognitive Sciences, 7, 337–347.
Ahn, H.-n., & Wampold, B. E. (2001). Where oh where are the specific ingredients? A meta-analysis of component studies in counseling and psychotherapy. Journal of Counseling Psychology, 48, 251–257.
American Psychiatric Association. (2013). Diagnostic and Statistical Manual of Mental Disorders (5th ed.). Arlington, VA: American Psychiatric Publishing.
Bamelis, L. L. M., Evers, S. M. A. A., Spinhoven, P., & Arntz, A. (2014). Results of a multicenter randomized controlled trial of the clinical effectiveness of schema therapy for personality disorders. American Journal of Psychiatry, 171, 305–322.
Beauchaine, T. P. (2001). Vagal tone, development, and Gray’s motivational theory: Toward an integrated model of autonomic nervous system functioning in psychopathology. Development and Psychopathology, 13, 183–214.
Beauchaine, T. P. (2015). Future directions in emotion dysregulation and youth psychopathology. Journal of Clinical Child and Adolescent Psychology, 44, 875–896.
Beauchaine, T. P., Klein, D. N., Crowell, S. E., Derbidge, C., & Gatzke-Kopp, L. (2009). Multifinality in the development of personality disorders: A biology × sex × environment interaction model of antisocial and borderline traits. Development and Psychopathology, 21, 735–770.
Beauchaine, T. P., & McNulty, T. (2013). Comorbidities and continuities as ontogenic processes: Toward a developmental spectrum model of externalizing psychopathology. Development and Psychopathology, 25, 1505–1528.
Belsky, D. W., Caspi, A., Arseneault, L., Bleidorn, W., Fonagy, P., Goodman, M., … Moffitt, T. E. (2012). Etiological features of borderline personality related characteristics in a birth cohort of 12-year-old children. Development and Psychopathology, 24, 251–265.
Bentler, P. M. (1980). Multivariate analysis with latent variables: Causal modeling. Annual Review of Psychology, 31, 419–456.
Bird, A. P. (1986). CpG-rich islands and the function of DNA methylation. Nature, 321, 209–213.
Bird, A. P. (2002). DNA methylation patterns and epigenetic memory. Genes and Development, 16, 6–21.
Bornovalova, M. A., Gratz, K. L., Delaney-Brumsey, A., Paulson, A., & Lejuez, C. W. (2006). Temperamental and environmental risk factors for borderline personality disorder among inner-city substance users in residential treatment. Journal of Personality Disorders, 20, 218–231.
Bornovalova, M. A., Hicks, B. M., Iacono, W. G., & McGue, M. (2009). Stability, change, and heritability of borderline personality disorder traits from adolescence to adulthood: A longitudinal twin study. Development and Psychopathology, 21, 1335–1353.
Bornovalova, M. A., Huibregtse, B. M., Hicks, B. M., Keyes, M., McGue, M., & Iacono, W. (2013). Tests of a direct effect of childhood abuse on adult borderline personality disorder traits: A longitudinal discordant twin design. Journal of Abnormal Psychology, 122, 180–194.
Bornovalova, M. A., Lejuez, C. W., Daughters, S. B., Rosenthal, M. Z., & Lynch, T. R. (2005). Impulsivity as a common process across borderline personality and substance use disorders. Clinical Psychology Review, 25, 790–812.
Bornovalova, M. A., Verhulst, B., Webber, T., McGue, M., Iacono, W. G., & Hicks, B. M. (2018). Genetic and environmental influences on the codevelopment among borderline personality disorder traits, major depression symptoms, and substance use disorder symptoms from adolescence to young adulthood. Development and Psychopathology, 30, 49–65.
Brewin, C. R., Andrews, B., & Valentine, J. D. (2000). Meta-analysis of risk factors for posttraumatic stress disorder in trauma-exposed adults. Journal of Consulting and Clinical Psychology, 68, 748–766.
Bridgett, D. J., Burt, N. M., Edwards, E. S., & Deater-Deckard, K. (2015). Intergenerational transmission of self-regulation: A multidisciplinary review and integrative conceptual framework. Psychological Bulletin, 141, 602–654.
Bromet, E. J., Atwoli, L., Kawakami, N., & Navarro-Mateu, F. (2017). Post-traumatic stress disorder associated with natural and human-made disasters in the World Mental Health Surveys. Psychological Medicine, 47, 227–241.
Brunner, R., Henze, R., Richter, J., & Kaess, M. (2015). Neurobiological findings in youth with borderline personality disorder. Scandinavian Journal of Child and Adolescent Psychiatry and Psychology, 3, 22–30.
Butler, T. R., Ariwodola, O. J., & Weiner, J. L. (2014). The impact of social isolation on HPA axis function, anxiety-like behaviors, and ethanol-drinking. Frontiers in Integrative Neuroscience, 7, 102–112.
Butler, T. R., Karkhanis, A. N., Jones, S. R., & Weiner, J. L. (2016). Adolescent social isolation as a model of heightened vulnerability to comorbid alcoholism and anxiety disorders. Alcoholism: Clinical and Experimental Research, 40, 1202–1214.
Canli, T., Ferra, J., & Duman, E. A. (2009). Genetics of emotion regulation. Neuroscience, 164, 43–54.
Caspi, A., Houts, R. M., Belsky, D. W., Goldman-Mellor, S. J., Harrington, H., Isreal, S., … Moffitt, T. E. (2014). The p factor: One general psychopathology factor in the structure of psychiatric disorders? Clinical Psychological Science, 2, 119–137.
Caspi, A., McClay, J., Moffitt, T. E., Mill, J., Martin, J., Craig, I. W., … Poulton, R. (2002). Role of genotype in the cycle of violence in maltreated children. Science, 297, 851–854.
Caspi, A., Taylor, A., Moffitt, T. E., & Plomin, R. (2000). Neighborhood deprivation affects children’s mental health: Environmental risks identified in a genetic design. Psychological Science, 11, 338–342.
Chanen, A. M., & Kaess, M. (2012). Developmental pathways to borderline personality disorder. Current Psychiatry Reports, 14, 45–53.
Checknita, D., Maussion, G., Labonté, B., Comai, S., Tremblay, R. E., Vitaro, F., … Turecki, G. (2015). Monoamine oxidase A gene promoter methylation and transcriptional downregulation in an offender population with antisocial personality disorder. British Journal of Psychiatry, 206, 216–222.
Chen, H., Cohen, P., Johnson, J. G., Kasen, S., Sneed, J. R., & Crawford, T. N. (2004). Adolescent personality disorders and conflict with romantic partners during the transition to adulthood. Journal of Personality Disorders, 18, 507–525.
Clark, A. T., Ye, H., Isbell, F., Deyle, E. R., Cowles, J., Tilman, G., & Sugihara, G. (2015). Spatial convergent cross mapping to detect causal relationships from short time series. Ecology, 96, 1174–1181.
Clarkin, J. F., Levy, K. N., Lenzenweger, M. F., & Kernberg, O. F. (2007). Evaluating three treatments of borderline personality disorder: A multiwave study. American Journal of Psychiatry, 164, 922–928.
Cohen, P., Chen, H., Gordon, K., Johnson, J., Brook, B., & Kasen, S. (2008). Socioeconomic background and the developmental course of schizotypal and borderline personality disorder symptoms. Development and Psychopathology, 20, 633–650.
Conway, C. C., Hammen, C., & Brennan, P. A. (2015). Adolescent precursors of adult borderline personality pathology in a high-risk community sample. Journal of Personality Disorders, 29, 316–333.
Crawford, T. N., Cohen, P. R., Chen, H., Anglin, D. M., & Ehrensaft, M. (2009). Early maternal separation and the trajectory of borderline personality disorder symptoms. Development and Psychopathology, 21, 1013–1030.
Crowell, S. E. (2016). Biting the hand that feeds: Current opinion on the interpersonal causes, correlates, and consequences of borderline personality disorder. F1000 Research, 5, 2796–2803.
Crowell, S. E., Baucom, B. R., McCauley, E., Potapova, N. V., Fitelson, M., Barth, H., … Beauchaine, T. P. (2013). Mechanisms of contextual risk for adolescent self-injury: Invalidation and conflict escalation in mother–child interactions. Journal of Clinical Child and Adolescent Psychology, 42, 467–480.
Crowell, S. E., Baucom, B. R., Yaptangco, M., Bride, D., Hsiao, R., McCauley, E., & Beauchaine, T. P. (2014). Emotion dysregulation and dyadic conflict in depressed and typical adolescents: Evaluating concordance across psychophysiological and observational measures. Biological Psychology, 98, 50–58.
Crowell, S. E., Beauchaine, T. P., Hsiao, R. C., Vasilev, C. A., Yaptangco, M., Linehan, M. M., & McCauley, E. (2012). Differentiating adolescent self-injury from adolescent depression: Possible implications for borderline personality development. Journal of Abnormal Child Psychology, 40, 45–57.
Crowell, S. E., Beauchaine, T. P., & Linehan, M. (2009). A biosocial developmental model of borderline personality: Elaborating and extending Linehan’s theory. Psychological Bulletin, 135, 495–510.
Crowell, S. E., Beauchaine, T. P., McCauley, E., Smith, C. J., Stevens, A. L., & Sylvers, P. (2005). Psychological, autonomic, and serotonergic correlates of parasuicide among adolescent girls. Development and Psychopathology, 17, 1105–1127.
Crowell, S. E., Butner, J. E., Wiltshire, T. J., Munion, A. K., Yaptangco, M., & Beauchaine, T. P. (2017). Evaluating emotional and biological sensitivity to maternal behavior among self-injuring and depressed adolescent girls using nonlinear dynamics. Clinical Psychological Science, 5, 272–285.
Crowell, S. E., & Kaufman, E. A. (2016). Borderline personality disorder and the emerging field of developmental neuroscience. Personality Disorders: Theory, Research, and Treatment, 7, 324–333.
Crowell, S. E., Yaptangco, M., & Turner, S. L. (2016). Coercion, invalidation, and risk for self-injury and borderline personality traits. In T. J. Dishion & J. J. Snyder (Eds.), The Oxford Handbook of Coercive Relationship Dynamics (pp. 182–193). New York: Oxford University Press.
Cuthbert, B. N., & Insel, T. R. (2013). Toward the future of psychiatric diagnosis: The seven pillars of RDoC. BMC Medicine, 11, 126–133.
Dammann, G., Teschler, S., Haag, T., Altmüller, F., Tuczek, F., & Dammann, R. H. (2011). Increased DNA methylation of neuropsychiatric genes occurs in borderline personality disorder. Epigenetics, 6, 1454–1462.
De Fruyt, F., & De Clercq, B. (2014). Antecedents of personality disorder in childhood and adolescence: Toward an integrative developmental model. Annual Review of Clinical Psychology, 10, 449–476.
Deckel, A. W., Hesselbrock, V., & Bauer, L. (1996). Antisocial personality disorder, childhood delinquency, and frontal brain functioning: EEG and neuropsychological findings. Journal of Clinical Psychology, 52, 639–650.
Dishion, T. J., Duncan, T. E., Eddy, J. M., Fagot, B. I., & Fetrow, R. (1994). The world of parents and peers: Coercive exchanges and children’s social adaptation. Social Development, 3, 255–268.
Dishion, T. J., McCord, J., & Poulin, F. (1999). When interventions harm: Peer groups and problem behavior. American Psychologist, 54, 755–764.
Doll, A., Sorg, C., Manoliu, A., Wöller, A., Meng, C., Förstl, H., … Riedl, V. (2013). Shifted intrinsic connectivity of central executive and salience network in borderline personality disorder. Frontiers in Human Neuroscience, 7, 1–13.
Eme, R. F. (2007). Sex differences in child-onset, life-course-persistent conduct disorder: A review of biological influences. Clinical Psychology Reviews, 27, 607–627.
Ensink, K., Biberdzic, M., Normandin, L., & Clarkin, J. (2015). A developmental psychopathology and neurobiological model of borderline personality disorder in adolescence. Journal of Infant, Child, and Adolescent Psychotherapy, 14, 46–69.
Fervaha, G., & Remington, G. (2013). Neuroimaging findings in schizotypal personality disorder: A systematic review. Progress in Neuro-Psychopharmacology and Biological Psychiatry, 43, 96–107.
Freud, S., & Breuer, J. (1893). On the psychical mechanism of hysterical phenomena: A preliminary communication. In S. Freud Selected Papers on Hysteria and Other Psychoneuroses (Trans. A. A. Brill, 1912). New York: The Journal of Nervous and Mental Disease Publishing Company. Available at www.bartleby.com/280/1.html
Fruzzetti, A. E., Shenk, C., & Hoffman, P. D. (2005). Family interaction and the development of borderline personality disorder. Development and Psychopathology, 17, 1007–1030.
Garattini, S., Giacalone, E., & Valzelli, L. (1967). Isolation, aggressiveness and brain 5-hydroxytryptamine turnover. Journal of Pharmacy and Pharmacology, 19, 338–339.
Gartstein, M. A., & Skinner, M. K. (2018). Prenatal influences on temperament development: The role of environmental epigenetics. Development and Psychopathology, 30, 1269–1303.
Goodman, M., Mascitelli, K., & Triebwasser, J. (2013). The neurobiological basis of adolescent-onset borderline personality disorder. Journal of the Canadian Academy of Child and Adolescent Psychiatry, 22, 212–219.
Gratz, K. L., Bardeen, J. R., Levy, R., Dixon-Gordon, K. L., & Tull, M. T. (2015). Mechanisms of change in an emotion regulation group therapy for deliberate self-harm among women with borderline personality disorder. Behaviour Research and Therapy, 65, 29–35.
Hajcak, G., MacNamara, A., & Olvet, D. M. (2010). Event-related potentials, emotion, and emotion regulation: An integrative review. Developmental Neuropsychology, 35, 129–155.
Hallquist, M. N., Hipwell, A. E., & Stepp, S. D. (2015). Poor self-control and harsh punishment in childhood prospectively predict borderline personality symptoms in adolescent girls. Journal of Abnormal Psychology, 124, 549–564.
Hedström, P., & Ylikoski, P. (2010). Causal mechanisms in the social sciences. Annual Review of Sociology, 36, 49–67.
Heinrichs, M., von Dawans, B., & Domes, G. (2009). Oxytocin, vasopressin, and human social behavior. Frontiers in Neuroendocrinology, 30, 548–557.
Hostinar, C. E., Sullivan, R. M., & Gunnar, M. R. (2014). Psychobiological mechanisms underlying the social buffering of the hypothalamic-pituitary-adrenocortical axis: A review of animal models and human studies across development. Psychological Bulletin, 140, 256–282.
Hughes, A. E., Crowell, S. E., Uyeji, L., & Coan, J. A. (2012). A developmental neuroscience of borderline pathology: Emotion dysregulation and social baseline theory. Journal of Abnormal Child Psychology, 40, 21–33.
Ingoldsby, E. M., & Shaw, D. S. (2002). Neighborhood contextual factors and early-starting antisocial pathways. Clinical Child and Family Psychology Review, 5, 21–55.
Joreskog, K. G., Sorbom, D., & Magidson, J. (1979). Advances in Factor Analysis and Structural Equation Models. New York: University Press of America.
Kaufman, E. A., & Crowell, S. E. (2018). Biological and behavioral mechanisms of identity pathology development: An integrative review. Review of General Psychology, 22, 245–263.
Kazdin, A. E. (2007). Mediators and mechanisms of change in psychotherapy research. Annual Review of Clinical Psychology, 3, 1–27.
Kenna, G. A., Roder-Hanna, N., Leggio, L., Zywiak, W. H., Clifford, J., Edwards, S., … Swift, R. M. (2012). Association of the 5-HTT gene-linked promoter region (5-HTTLPR) polymorphism with psychiatric disorders: Review of psychopathology and pharmacotherapy. Pharmacogenomics and Personalized Medicine, 5, 19–35.
Knoblich, N., Gundel, F., Brückmann, C., Becker-Sadzio, J., Frischholz, C., & Nieratschker, V. (2018). DNA methylation of APBA3 and MCF2 in borderline personality disorder: Potential biomarkers for response to psychotherapy. European Neuropsychopharmacology, 28, 252–263.
Koenig, J., Kemp, A. H., Feeling, N. R., Thayer, J. F., & Kaess, M. (2016). Resting state vagal tone in borderline personality disorder: A meta-analysis. Progress in Neuro-Psychopharmacology and Biological Psychiatry, 64, 18–26.
Kolla, N. J., Chiuccariello, L., Wilson, A. A., Houle, S., Links, P., Bagby, R. M., … Meyer, J. H. (2016). Elevated monoamine oxidase-A distribution volume in borderline personality disorder is associated with severity across mood symptoms, suicidality, and cognition. Biological Psychiatry, 79, 117–126.
Kotov, R., Krueger, R. F., Watson, D., Achenbach, T. M., Althoff, R. R., Bagby, R. M., … Zimmerman, M. (2017). The Hierarchical Taxonomy of Psychopathology (HiTOP): A dimensional alternative to traditional nosologies. Journal of Abnormal Psychology, 126, 454–477.
Krueger, R. F., & Markon, K. E. (2014). The role of the DSM-5 personality trait model in moving toward a quantitative and empirically based approach to classifying personality and psychopathology. Annual Review of Clinical Psychology, 10, 477–501.
Kumsta, R., Hummel, E., Chen, F. S., & Heinrichs, M. (2013). Epigenetic regulation of the oxytocin receptor gene: Implications for behavioral neuroscience. Frontiers in Neuroscience, 7, 1–6.
Kupfer, D. J., First, M. B., & Regier, D. A. (2002). Introduction. In D. J. Kupfer, M. B. First, & D. A. Regier (Eds.), A Research Agenda for DSM-V (pp. xv–xxiii). Washington, DC: American Psychiatric Association.
Larsson, H., Andershed, H., & Lichtenstein, P. (2006). A genetic factor explains most of the variation in the psychopathic personality. Journal of Abnormal Psychology, 115, 221–230.
Lawrence, K. A., Allen, J. S., & Chanen, A. M. (2010). Impulsivity in borderline personality disorder: Reward-based decision-making and its relationship to emotional distress. Journal of Personality Disorders, 24, 785–799.
Lee, R. (2006). Childhood trauma and personality disorder: Toward a biological model. Current Psychiatry Reports, 8, 43–52.
Lejuez, C. W., Daughters, S. B., Nowak, J. A., Lynch, T., Rosenthal, M. Z., & Kosson, D. (2003). Examining the inventory of interpersonal problems as a tool for conducting analogue studies of mechanisms underlying borderline personality disorder. Journal of Behavior Therapy and Experimental Psychiatry, 34, 313–324.
Lieb, K., Zanarini, M. C., Schmahl, C., Linehan, M. M., & Bohus, M. (2004). Borderline personality disorder. Lancet, 364, 453–461.
Linehan, M. (1993). Cognitive-Behavioral Treatment of Borderline Personality Disorder. New York: Guilford Press.
Lowry, P. B., & Gaskin, J. (2014). Partial least squares (PLS) structural equation modeling (SEM) for building and testing behavioral causal theory: When to choose it and how to use it. IEEE Transactions on Professional Communication, 57, 123–146.
Lynam, D. R., Caspi, A., Moffitt, T. E., Wikström, P. H., Loeber, R., & Novak, S. (2000). The interaction between impulsivity and neighborhood context on offending: The effects of impulsivity are stronger in poorer neighborhoods. Journal of Abnormal Psychology, 109, 563–574.
Lyons-Ruth, K. (2008). Contributions of the mother–infant relationship to dissociative, borderline, and conduct symptoms in young adulthood. Infant Mental Health Journal, 29, 203–218.
Lyons-Ruth, K., Bureau, J., Easterbrooks, M. A., Obsuth, I., & Hennighausen, K. (2013). Parsing the construct of maternal insensitivity: Distinct longitudinal pathways associated with early maternal withdrawal. Attachment and Human Development, 15, 562–582.
Lyons-Ruth, K., Bureau, J., Holmes, B., Easterbrooks, A., & Brooks, N. H. (2013). Borderline symptoms and suicidality/self-injury in late adolescence: Prospectively observed relationship correlates in infancy and childhood. Psychiatry Research, 206, 273–281.
MacKinnon, D. F., & Pies, R. (2006). Affective instability as rapid cycling: Theoretical and clinical implications for borderline personality and bipolar spectrum disorders. Bipolar Disorders, 8, 1–14.
Mahoney, J. (2001). Beyond correlational analysis: Recent innovations in theory and method. Sociological Forum, 16, 575–593.
Markon, K. E., & Jonas, K. G. (2016). Structure as cause and representation: Implications of descriptivist inference for structural modeling across multiple levels of analysis. Journal of Abnormal Psychology, 125, 1146–1157
Martin, E. I., Ressler, K. J., Binder, E., & Nemeroff, C. B. (2010). The neurobiology of anxiety disorders: Brain imaging, genetics, and psychoneuroendocrinology. Clinics in Laboratory Medicine, 30, 865–891.
McLaughlin, K. A., & Lambert, H. K. (2017). Child trauma exposure and psychopathology: Mechanisms of risk and resilience. Current Opinion in Psychology, 14, 29–34.
Miller, J. D., Lyman, D. R., Widiger, T. A., & Leukefeld, C. (2001). Personality disorders as extreme variants of common personality dimensions: Can the five factor model adequately represent psychopathy? Journal of Personality, 69, 253–276.
Mitchell, A. E., Dickens, G. L., & Picchioni, M. M. (2014). Facial emotion processing in borderline personality disorder: A systematic review and meta-analysis. Neuropsychology Review, 24, 166–184.
Monk, C., Spicer, J., & Champagne, F. A. (2012). Linking prenatal maternal adversity to developmental outcomes in infants: The role of epigenetic pathways. Development and Psychopathology, 24, 1361–1376.
Moul, C., Killcross, S., & Dadds, M. R. (2012). A model of differential amygdala activation in psychopathy. Psychological Review, 119, 789–806.
Mushtaq, R., Shoib, S., Shah, T., & Mushtaq, S. (2014). Relationship between loneliness, psychiatric disorders and physical health: A review on the psychological aspects of loneliness. Journal of Clinical and Diagnostic Research, 8, 1–4.
Myers, C. S. (1915). A contribution to the study of shell shock: Being an account of three cases of loss of memory, vision, smell, and taste, admitted into the Duchess of Westminster’s war hospital, Le Touquet. Lancet, 185, 316–320.
Nelson, S. E., & Dishion, T. J. (2004). From boys to men: Predicting adult adaptation from middle childhood sociometric status. Development and Psychopathology, 16, 441–459.
Neuhaus, E., & Beauchaine, T. P. (2017). Impulsivity and vulnerability to psychopathology. In T. P. Beauchaine & S. P. Hinshaw (Eds.), Child and Adolescent Psychopathology (pp. 178–212). Hoboken, NJ: John Wiley.
Oberlander, T. F., Weinberg, J., Papsdorf, M., Grunau, R., Misri, S., & Devlin, A. M. (2008). Prenatal exposure to maternal depression, neonatal methylation of human glucocorticoid receptor gene (NR3C1) and infant cortisol stress responses. Epigenetics, 3, 97–106.
Pagliaccio, D., Luby, J. L., Bogdan, R., Agrawal, A., Gaffrey, M. S., Belden, A. C., … Barch, D. M. (2015). Amygdala functional connectivity, HPA axis genetic variation, and life stress in children and relations to anxiety and emotion regulation. Journal of Abnormal Psychology, 124, 817–833.
Patterson, G. R. (1982). Coercive Family Processes (Vol. 3). Eugene, OR: Castalia Publishing Company.
Patterson, G. R., DeBaryshe, B., & Ramsey, E. (1990). A developmental perspective on antisocial behavior. American Psychologist, 44, 329–335.
Perroud, N., Paoloni-Giacobino, A., Prada, P., Olié, E., Salzmann, A., Nicastro, R., … Malafosse, A. (2011). Increased methylation of glucocorticoid receptor gene (NR3C1) in adults with a history of childhood maltreatment: A link with the severity and type of trauma. Translational Psychiatry, 1, 1–9.
Perroud, N., Salzmann, A., Prada, P., Nicastro, R., Hoeppli, M-E., Furrer, S., … Malafosse, A. (2013). Response to psychotherapy in borderline personality disorder and methylation status of the BDNF gene. Translational Psychiatry, 3, e207–e214.
Piehler, T. F., & Dishion, T. J. (2008). Interpersonal dynamics within adolescent friendships: Dyadic mutuality, deviant talk, and patterns of antisocial behavior. Child Development, 78, 1611–1624.
Posner, M. I., Rothbart, M. K., Vizueta, N., Levy, K. N., Evans, D. E., Thomas, K. M., & Clarkin, J. F. (2002). Attentional mechanisms of borderline personality disorder. Proceedings of the National Academy of Sciences USA, 99, 16366–16370.
Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, Instruments, and Computers, 36, 717–731.
Putnam, K. M., & Silk, K. R. (2005). Emotion dysregulation and the development of borderline personality disorder. Development and Psychopathology, 17, 899–925.
Raine, A., Venables, P. H., & Mednick, S. A. (1997). Low resting heart rate at age 3 years predisposes to aggression at age 11 years: Evidence from the Mauritius Child Health Project. Journal of the American Academy of Child and Adolescent Psychiatry, 36, 1457–1464.
Saul, L. J. (1945). Psychological factors in combat fatigue: With special reference to hostility and the nightmares. Psychosomatic Medicine, 7, 257–272.
Shalev, A., Liberzon, I., & Marmar, C. (2017). Post-traumatic stress disorder. The New England Journal of Medicine, 376, 2459–2469.
Sharma, L., Markon, K. E., & Clark, L. A. (2014). Toward a theory of distinct types of “impulsive” behaviors: A meta-analysis of self-report and behavioral measures. Psychological Bulletin, 140, 374–408.
Sharp, C., & Romero, C. (2007). Borderline personality disorder: A comparison between children and adults. Bulletin of the Menninger Clinic, 71, 85–114.
Skodol, A. E., Siever, L. J., Livesley, W. J., Gunderson, J. G., Pfohl, B., & Widiger, T. A. (2002). The borderline diagnosis II: Biology, genetics, and clinical course. Biological Psychiatry, 51, 951–963.
Snyder, J., Schrepferman, L., McEachern, A., Barner, S., Johnson, K., & Provines, J. (2008). Peer deviancy training and peer coercion: Dual processes associated with early-onset conduct problems. Child Development, 79, 252–268.
Snyder, J., Schrepferman, L., & St. Peter, C. (1997). Origins of antisocial Siever, L. (2010). An attachment perspective on borderline personality disorder: Advances in gene–environment considerations. Current Psychiatry Reports, 12, 61–67.
Stephan, K. E., Penny, W. D., Moran, R. J., den Ouden, H. E. M., Daunizeau, J., & Friston, K. J. (2010). Ten simple rules for dynamic causal modeling. NeuroImage, 49, 3099–3109.
Stepp, S. D., Lazarus, S. A., & Byrd, A. L. (2015). A systematic review of risk factors prospectively associated with borderline personality disorder: Taking stock and moving forward. Personality Disorders: Theory, Research, and Treatment, 7, 316–323.
Stepp, S. D., Scott, L. N., Jones, N. P., Whalen, D. J., & Hipwell., A. P. (2016). Negative emotional reactivity as a marker of vulnerability in the development of borderline personality disorder symptoms. Development and Psychopathology, 28, 213–224.
Stetler, C., & Miller, G. E. (2011). Depression and hypothalamic-pituitary-adrenal activation: A quantitative summary of four decades of research. Psychosomatic Medicine, 73, 114–126.
Strauman, T. J. (2017). Self-regulation and psychopathology: Toward an integrative translational paradigm. Annual Review of Clinical Psychology, 13, 497–523.
Susman, E. J. (2006). Psychobiology of persistent antisocial Sternberg, E. (2006). Beyond heart rate variability: Vagal regulation of allostatic systems. Annals of the New York Academy of Sciences, 1088, 361–372.
Trull, T. J., Stepp, S. D., & Durrett, C. A. (2003). Research on borderline personality disorder: An update. Current Opinion in Psychiatry, 16, 77–82.
Tyrer, P., Reed, G. M., & Crawford, M. J. (2015). Classification, assessment, prevalence, and effect of personality disorder. Lancet, 385, 21–27.
van Zutphen, L., Siep, N., Jacob, G. A., Goebel, R., & Arntz, A. (2015). Emotional sensitivity, emotion regulation and impulsivity in borderline personality disorder: A critical review of fMRI studies. Neuroscience and Biobehavioral Reviews, 51, 64–76.
Viding, E., Blair, R. J. R., Moffitt, T. E., & Plomin, R. (2005). Evidence for substantial genetic risk for psychopathy in 7-year-olds. Journal of Child Psychology and Psychiatry, 46, 592–597.
Widiger, T. A., & Simonsen, E. (2005). Alternative dimensional models of personality disorder: Finding a common ground. Journal of Personality Disorders, 19, 110–130.
Winsper, C., Marwaha, S., Lereya, S. T., Thompson, A., Eyden, J., & Singh, S. P. (2016). A systematic review of the neurobiological underpinnings of borderline personality disorder (BPD) in childhood and adolescence. Reviews in the Neurosciences, 27, 827–847.
Wright, A. G. C., Thomas, K. M., Hopwood, C. J., Markon, K. E., Pincus, A. L., & Krueger, R. F. (2012). The hierarchical structure of DSM-5 pathological personality traits. Journal of Abnormal Psychology, 121, 951–957.
Zuckerman, M., & Kuhlman, M. (2000). Personality and risk-tasking: Common biosocial factors. Journal of Personality, 68, 999–1029.