Emotional dysregulation is a feature of many, if not most, forms of psychopathology, and is often considered a precursor in the development of disorder (Berenbaum, Raghavan, Le, Vernon, & Gomez, Reference Berenbaum, Raghavan, Le, Vernon and Gomez2003; Bradley, Reference Bradley2003; Cicchetti, Ackerman, & Izard, Reference Cicchetti, Ackerman and Izard1995; Cole, Michel, & Teti, Reference Cole, Michel and Teti1994; Gross & Muñoz, Reference Gross and Muñoz1995; Keenan, Reference Keenan2000; Kring & Werner, Reference Kring, Werner, Philippot and Feldman2004). In the past few decades, there has been enormous scientific progress toward understanding the development of emotion regulation, its biological and behavioral aspects, and individual differences that constitute risk for psychological disorder. A broad range of studies indicates that higher levels of negative emotion, whether conceptualized as inherited disposition or as acquired response patterns, are associated with risk for a host of disorders. There is, however, a general lack of specificity in our predictions of particular clinical outcomes. To advance our understanding, and to inform early identification and prevention, it would be valuable to increase the specificity with which we trace developmental pathways between early childhood emotional functioning and later emerging symptoms and disorders. One strategy that may aid this effort is to more precisely focus on the ways that emotional functioning is dysregulated in different disorders (other than general negative emotionality), and to link those patterns to earlier emotional functioning, attending to the emotional characteristics of environmental risk conditions.
In this article, we consider the developmental links between the emotional sequelae of young children's exposure to risk environments, particularly parental maltreatment, and of individual differences in children's sensitivity to them, that are thought to play a role in the development of the emotional dysregulation associated with borderline personality disorder (BPD). We first discuss emotional instability and poor emotional awareness as key features of emotional dysregulation in borderline personality. Next, we briefly summarize views of how a child's dispositional vulnerabilities and exposure to emotionally overwhelming family circumstances and inadequate caregiving contribute to the emotional development of borderline personality. We then attempt to link evidence of the effects of these risk factors on early childhood emotional functioning with the emotional dysregulation associated with borderline personality, revealing a need for research to fill the gaps in our knowledge of these relations.
We offer a developmental psychopathology approach to the design of that research. That is, we appreciate the transactional nature of development (e.g., the complex and dynamic interplay of child individual characteristics, of family factors, and of the contexts in which families develop), the need for multiple levels of analysis (e.g., biological, behavioral, and experiential), and the principles that different pathways can lead to a similar outcome (different risk factors can lead to borderline personality) and that initial common circumstances can lead to diverse outcomes (a history of child sexual abuse can lead to a variety of adult outcomes, not just borderline personality; Goodman & Yehuda, Reference Goodman and Yehuda2002; Paris, Reference Paris1998). Moreover, we advocate for integrated experimental and prospective longitudinal studies, particularly those that focus on early childhood when risk for BPD may be greatest, to increase the specificity with which we understand any of these pathways. Our effort leads us to emphasize the importance of conceptualizing the proximal emotional environment of the young at risk child.
Emotional Dysregulation as a Feature of BPD
Despite vigorous debate about the existence and nature of the “borderline” personality, it is a concept that has had clinical staying power. Stern (Reference Stern1938) is believed to have introduced the term as he struggled to conceptualize clients who did not appear to suffer thought disorder but who nonetheless demonstrated remarkably poor reality testing and inexplicably destructive behavior at times. These individuals seemed to be on the borderline between neurosis and psychosis. That is, Stern searched for a way to understand how individuals who seemed to function relatively well in many aspects of their lives could at times behave erratically, with poor reality testing, and little insight into their emotional reactions. To us, what is essential about the origin of the term borderline, albeit scientifically dissatisfying, is its emphasis on marked variability in and impoverished self-understanding of emotional functioning. That is, symptomatic behavior in individuals with BPD is related to their perceptions that trigger intense emotional shifts without their having insight into the reasons for such reactions. When their emotional buttons are pushed, individuals who function relatively well at other times behave in markedly problematic ways. Their emotional shifts seem exaggerated and unpredictable, and they are unaware of or confused by them; they lack insight into what motivates their intense reactions, often fail to achieve a shared perception of reality about the circumstances with those most close to them, and engage in behaviors that destroy their relationships and even their lives.
The effort to understand the nature and origins of BPD has fostered much debate and deep thought. It is undecided whether there is one core cluster of symptoms or whether BPD is comprised of multiple factors and subtypes (e.g., Lenzenweger, Clarkin, Yeomans, Kernberg, & Levy, Reference Lenzenweger, Clarkin, Yeomans, Kernberg and Levy2008; Paris, Reference Paris2007; Sanislow et al., Reference Sanislow, Grilo, Morey, Bender, Skodol and Gunderson2002). It is debated whether BPD is a discrete category or a deviation in the continuum of ego or personality functioning (Cloninger, Svrakic, & Przybeck, 1993; Kernberg, Reference Kernberg1975; Skodol, Gunderson, Pfohl, Widiger, Livesley, & Siever, Reference Skodol, Gunderson, Pfohl, Widiger, Livesley and Siever2002; Trull & Durrett, Reference Trull and Durrett2005). In terms of its origins, it has been variously conceptualized as a product of specific types of family-based trauma (Herman, Russell, & Trocki, Reference Herman, Russell and Trocki1986), as a product of genetic predispositions (harm avoidance and novelty seeking; Svrakic, Whitehead, Przybeck, & Cloninger, Reference Svrakic, Whitehead, Przybeck and Cloninger1993), and as a diathesis of biologically based vulnerabilities (emotional reactivity, impulsivity, interpersonal sensitivity) and exposure to family environments (invalidating, abusive) that compromise children's socioemotional learning, attachment security, and neurophysiology (Crowell, Beauchaine, & Linehan, Reference Crowell, Beauchaine and Linehan2009; Fonagy, Gergely, Jurist, & Target, Reference Fonagy, Gergely, Jurist and Target2002; Gunderson & Ruth-Lyons, 2008; Linehan, Reference Linehan1993; Putnam & Silk, Reference Putnam and Silk2005). Despite these unresolved issues, all of these views refer to emotional dysregulation as a core aspect of BPD. Defined somewhat variously across models, there is convergence on two unique features of the emotional dysregulation of borderline personality: emotional instability (sudden, unexpected, intense emotional reactions) and poor emotional awareness (including dissociation, denial, emotional avoidance, and emotional cutting off).
Emotional instability refers to unexpected, intense, and rapid changes in emotion (Siever & Davis, Reference Siever and Davis1991). Although emotional reactivity is often the focus of discussion about borderline personality dysfunction (e.g., Linehan, Reference Linehan1993; Livesley, Jackson, & Schroeder, Reference Livesley, Jackson and Schroeder1992), emotional instability may distinguish BPD from other disorders that involve emotional reactivity (e.g., oppositional defiant disorder, intermittent explosive disorder). That is, seeming fine one moment, then intensely angry without apparent reason and later indifferent may be particularly distinctive of borderline personality. In contrast to bipolar disorder, in which shifts in mood appear because of biological dysregulation, borderline emotional instability may be more context dependent, particularly to interpersonal context (Gunderson, Reference Gunderson2007). Although the evidence remains to be established, clinicians believe that emotional instability in BPD is often associated with implicit perceptions that another is becoming too close or has lapsed in attentiveness. The individual with BPD may not over react to a broken soda machine but may be highly reactive when perceiving a hint of loss of attention or when feeling engulfed by attention or love.
Research on the emotional functioning of individuals with BPD has focused more on reactivity than instability until recently. Most individuals with BPD describe themselves as emotionally unstable, and to a greater extent than those with other personality disorders (Siever, Torgersen, Gunderson, Livesley, & Kendler, Reference Siever, Torgersen, Gunderson, Livesley and Kendler2002; Zanarini, Frankenburg, Hennen, & Silk, Reference Zanarini, Frankenburg, Hennen and Silk2003). A 2-year longitudinal study found that prior emotional instability, more so than impulsivity, predicted later BPD symptoms in young adults (Tragesser, Solhan, Schwartz-Mette, & Trull, Reference Tragesser, Solhan, Schwartz-Mette and Trull2007). Also, individuals with BPD reported larger changes in emotions over a day than healthy controls (Ebner-Priemer et al., Reference Ebner-Priemer, Kuo, Kleindeinst, Welch, Reisch and Reinhard2007) and over a month compared to depressed controls (Trull et al., Reference Trull, Solhan, Tragesser, Jahng, Wood and Piasecki2008), a high degree of intraindividual variability in pleasant and unpleasant emotions (Russell, Moskowitz, Zuroff, Sookman, & Paris, Reference Russell, Moskowitz, Zuroff, Sookman and Paris2007), and a pattern of fluctuations between anxiety and depression or anger (Koenigsberg et al., Reference Koenigsberg, Harvey, Mitropoulou, Schmeider, New and Goodman2002).
Whereas typical adults generally appear to have relatively smooth, subtle, and predictable transitions from appearing affectively neutral to responding emotionally in interpersonal situations, the person with BPD has unexpected, unpredictable changes in emotion. Moreover, even when the typical adult does react intensely to a situation, it is usually appropriate to the context. The emotional reactions of individuals with BPD appear exaggerated or inexplicable in the circumstances. For instance, a hospitalized 15-year-old was eager for a family therapy session for which she had been preparing for weeks in individual therapy. At last, she had earned the privilege of a weekend pass. As she met her family at the door of the therapy room, the girl erupted in a rage that no one understood. Unable to control her rage, she had to be carried to an isolation room, where she remained for hours. Eventually it emerged that her rage was triggered by seeing that her mother had been wearing the daughter's belt. This sight generated rage that, with reflection, appeared because of the mother's self-centeredness, intrusiveness, and indifference to her daughter's needs. These types of anecdotes suggest the importance of studying the temporal dynamics of emotional responses, their context inappropriateness, and their role in inappropriate behavior (Cole & Hall, Reference Cole, Hall, Beauchaine and Hinshaw2008); these distinguish emotional instability from emotional reactivity of other disorders and normal mood fluctuations of highly expressive individuals.
Poor emotional awareness is less often studied in research on borderline personality but it seems critical to understanding the unique features of emotional dysregulation in this disorder. A number of different but related phenomena can be considered in this aspect of emotional dysregulation including emotional confusion, cutting off, avoidance, denial, and dissociation. One view is that poor emotional awareness is at the core of borderline personality and underlies emotional instability (Farrell & Shaw, Reference Farrell and Shaw1994). Inspired by the focus on emotion in clinical work espoused by Greenberg and Safran (Reference Greenberg and Safran1984), Farrell and Shaw suggested that emotional awareness provides feedback to the self about emotions that then facilitates better regulation of emotions. This self-reflective process has also been referred to as mentalization or reflective function (Fonagy et al., Reference Fonagy, Gergely, Jurist and Target2002). These views underscore the disadvantages of being unable to regard internal states as objects of thought; emotions and the situations that evoke them are harder to understand and resolve in socially appropriate ways. In our view, the inability to realistically appraise one's reactions and their situational contexts, the central skill that is taught to young children in emotional competence promoting prevention programs (e.g., Greenberg, Kusché, Cook, & Quamma, Reference Greenberg, Kusché, Cook and Quamma1995), not only impedes adaptive self-regulation but also interferes significantly with the ability of one's close relationships to serve regulatory functions. One of the most frustrating aspects of interacting with individuals with BPD is their refusal or inability to allow others to soothe them, even when demanding attention and solace. If one cannot reflect on internal states and their circumstances realistically, it is difficult to engage in a shared reality with loved ones and involve them in resolving these issues.
The evidence illustrates the complexity of poor emotional awareness among individuals with borderline personality. Persons diagnosed with BPD demonstrate poorer emotional awareness (e.g., Levine, Marziali, & Hood, Reference Levine, Marziali and Hood1997; Wolff, Stiglmayr, Bretz, Lammers, & Auckenthaler, Reference Wolff, Stiglmayr, Bretz, Lammers and Auckenthaler2007) and yet they appear overly sensitive to emotional stimuli (e.g., Domes et al., Reference Domes, Czieschnek, Weidler, Berger, Fast and Herpertz2008; Donegan et al., Reference Donegan, Sanislow, Blumberg, Fulbright, Lacadie and Skudlarski2003; Lynch et al., Reference Lynch, Rosenthal, Kosson, Cheavens, Lejuez and Blair2006; Wagner & Linehan, Reference Wagner and Linehan1999). Possibly, they are overly vigilant in regard to how others are behaving, often perceiving neutral interpersonal cues as threatening, which destabilizes them emotionally, leaving them unable to reflect on their reactions once they are overwhelmed emotionally (e.g., Wolff et al., Reference Wolff, Stiglmayr, Bretz, Lammers and Auckenthaler2007). Their impoverished emotional awareness, therefore, may not reflect a global cognitive impairment but a chronic tendency for awareness to be diminished when stress, notably interpersonal stress, evokes strong emotional reactions (Gunderson, Reference Gunderson2007). Individuals with BPD report that interpersonal stressors trigger aversive tension (Stiglmayr et al., Reference Stiglmayr, Grathwol, Linehan, Ihorst, Fahrenberg and Bohus2005). When interpersonal stress overwhelms them emotionally, individuals with BPD may engage in cognitive processes that cloud emotional awareness. Dissociation, thought suppression, or cutting off from emotional experience are conceptualized as implicit mechanisms that serve to avoid overwhelming, intolerable emotions (Cheavens et al., Reference Cheavens, Rosenthal, Daughters, Nowak, Kosson and Lynch2005; Gratz, Tull, & Gunderson, Reference Gratz, Tull and Gunderson2008; Rosenthal, Cheavens, Lejuez, & Lynch, Reference Rosenthal, Cheavens, Lejuezb and Lynch2005). In addition, it has been suggested that an aspect of the “emotional cascade” in BPD is rumination, which contributes to being emotionally overwhelmed, and which ultimately (though temporarily) is only relieved by emotional avoidance (Selby, Anestis, Bender, & Joiner, Reference Selby, Anestis, Bender and Joiner2009). Individuals with BPD report significantly more frequent, intense, rapidly escalating, and longer lasting states of aversive tension compared to controls (Stiglmayr et al., Reference Stiglmayr, Grathwol, Linehan, Ihorst, Fahrenberg and Bohus2005), and engage in maladaptive behaviors to control or cope with tension (e.g., Brodsky, Groves, Oquendo, Mann, & Stanley, Reference Brodsky, Groves, Oquendo, Mann and Stanley2006; Gratz, Reference Gratz2003). Together, the data indicate that individuals with BPD experience more internal distress, and may only avoid this distress by transforming it into external pain or by suppressing thoughts that foster it. Avoidance-based coping provides temporary relief, but maintains psychological difficulties in the longer term (Hayes, Wilson, Gifford, Follette, & Strosahl, Reference Hayes, Wilson, Gifford, Follette and Strosahl1996). These findings suggest the importance of assessing the degree to which a child at risk for BPD is able to tolerate strong emotion, the cognitive and behavioral activities that follow strong emotion, and the capacity for recall of the evocative event.
Neurocognitive Underpinnings of Emotional Instability and Poor Emotional Awareness
A discussion of emotional instability and poor emotional awareness is incomplete without appreciating the neurocognitive functions that support emotion regulation, particularly in light of growing evidence that individuals with BPD differ from clinical and healthy controls in neurocognitive functioning. This aspect of impairment may moderate the development of the emotional dysregulation in BPD, contributing to other contributing factors such as insecure attachment (e.g., Judd, Reference Judd2005) and emotional reactivity (Fertuck, Lenzenweger, Clarkind, Hoermann, & Stanley, Reference Fertuck, Lenzenweger, Clarkind, Hoermann and Stanley2006; Ruocco, Reference Ruocco2005). Neurological impairment in individuals with BPD includes hyperresponsiveness of the noradrenergic and hypothalamic–pituitary–adrenal (HPA) systems (e.g., Crowell et al., Reference Crowell, Beauchaine and Linehan2009; Figueroa & Silk, Reference Figueroa and Silk1997; Siever & Davis, Reference Siever and Davis1991), which has implications for emotional instability. HPA hyperreactivity is greatest among individuals with BPD who also have a history of trauma (Jogems-Kosterman, de Knijff, Kusters, & van Hoof, Reference Jogems-Kosterman, de Knijff, Kusters and van Hoof2007; Rinne et al., Reference Rinne, De Kloet, Wouters, Goekoop, de Rijk and van den Brink2002). Dissociative symptoms are associated with greater HPA and noradrenergic stress reactivity among individuals with BPD or trauma history (Simeon, Knutelska, Smith, Baker, & Hollander, Reference Simeon, Knutelska, Smith, Baker and Hollander2007), which has implications for poor emotional awareness. Many clients with BPD report childhood trauma; it may be that trauma moderates associations between symptoms and physiological functioning through emotional instability and poor emotional awareness.
Structural brain differences between persons with BPD and controls also are found, involving areas of the amygdala, hippocampus, orbitofrontal cortex, and anterior cingulated cortex, all of which involve neurocircuitry that contributes to emotion regulation (see Grosjean & Tsai, Reference Grosjean and Tsai2007; New, Triebwasser, & Charney, Reference New, Triebwasser and Charney2008; Rosenthal et al., Reference Rosenthal, Gratz, Kosson, Cheavens, Lejuez and Lynch2008). In terms of neurotransmission, broad differences in serotonergic, dopaminergic, and glutamatergic circuits are also found, with implications for aggression, affective instability, impulsivity, and negative affectivity (see Crowell et al., Reference Crowell, Beauchaine and Linehan2009; Grosjean & Tsai, Reference Grosjean and Tsai2007). Structural abnormalities in adults with BPD may reflect long-term effects of early environmental stress and its cascading effects on functional neurotransmission over time. Grosjean and Tsai (Reference Grosjean and Tsai2007) point out that the N-methyl-d-aspartate system regulates a critical period of growth and metabolism in a wide range of corticolimbic areas and that early environmental stress compromises N-methyl-d-aspartate activity. Resulting cell death in areas that support attention control, which is needed for both emotion stability and awareness (Milham & Banich, Reference Milham and Banich2005), could influence the ability to store memories for stressful events, which would then affect emotional awareness.
In sum, emotional instability and poor emotional awareness are distinctive emotional symptoms that are associated with a host of other disabling BPD symptoms, including self-harm, suicidality, impulsivity, disturbed relationships, disturbed identity, and feelings of emptiness. Neurocognitive impairments may contribute to them, but this may depend upon specific qualities of exposure to early trauma. The seriousness of these problems for the individual and their loved ones underscores the importance of discerning the conditions that create risk for the development of the borderline personality.
Conditions of Risk Associated With the Development of Borderline Personality
From a developmental psychopathology framework, we adopt the view that there is no single pathway to borderline personality. We focus on a pathway in which conditions of risk interact in ways that contribute to the development of emotional instability and poor emotional awareness. Given the prevalent view that the origins of borderline functioning lie in early childhood, we turn now to evidence of the effects of early childhood risk on children's emotional functioning and development, with an eye toward evidence of continuity between early childhood emotional functioning when putative risks are present and the emotional instability and poor emotional awareness that characterizes BPD. To pursue these questions, we consider several leading views on the etiology of BPD and then move to the literature on the effects of those risk factors on young children's emotional functioning.
The common view of the development of BPD is that it arises from the interplay between a child's temperament and caregiving failures. Linehan (Reference Linehan1993), for instance, discusses the at-risk child as being emotionally reactive and the child's caregivers as invalidating the child's emotional experience via intense emotional exchanges that fail to teach the child emotion regulation strategies. In the reformulated model (Crowell et al., Reference Crowell, Beauchaine and Linehan2009), impulsivity is proposed as an additional risk factor attributed to temperament. Attachment-based models (Fonagy et al., Reference Fonagy, Gergely, Jurist and Target2002; Gunderson & Lyons-Ruth, Reference Gunderson and Lyons-Ruth2008) focus more on relationships than interactions alone, attributing the development of BPD to biologically based vulnerabilities that make children vulnerable to parental weaknesses that create challenges for the child's sense of security and self. Gunderson and Lyons-Ruth (Reference Gunderson and Lyons-Ruth2008) highlight the child's interpersonal sensitivity, appreciating that children who are biologically reactive to their circumstances may be particularly vulnerable to failures in caregiving and more likely to form an insecure attachment. Fonagy and colleagues focus in addition on the importance of parental guidance in developing the ability to mentalize or reflect on experiences of oneself and of others. All of these models illustrate the need to understand a dynamic, transactional interplay of child and environmental risk factors that in concert contribute to the emotional dysregulation associated with BPD.
Infant Temperament as a Risk Factor
Infant temperament, particularly negative emotionality in the presence of low effortful control, is associated with many forms of psychopathology (Rothbart & Bates, Reference Rothbart, Bates, Eisenberg, Damon and Lerner2006). Negative emotionality is regarded as a highly heritable dimension of child temperament and adult personality (e.g., Bouchard & Loehlin, Reference Bouchard and Loehlin2001; Goldsmith, Lemery, Buss, & Campos, Reference Goldsmith, Lemery, Buss and Campos1999), such that a child prone to negativity is also more likely to have a parent with that disposition. The evidence for biological risk from genetic studies indicates considerable heritability for neuroticism, negative emotionality, and emotional instability (e.g., Keller, Coventry, Heath, & Martin, Reference Keller, Coventry, Heath and Martin2005; Loehlin, Willerman, & Horn, Reference Loehlin, Willerman and Horn1985). However, the intricate ways that environment and gene expression influence each other must be considered. Early exposure to stress, both pre- and postnatally, may contribute to temperamental vulnerability in infancy (e.g., Buitelaar, Huizink, Mulder, de Medina, & Visser, Reference Buitelaar, Huizink, Mulder, de Medina and Visser2003; Pesonen et al., Reference Pesonen, Räikkönen, Heinonen, Komsi, Järvenpää and Strandberg2008).
Even for parents who do not have genetically based vulnerabilities, emotionally reactive infants are challenging to parents; they react intensely and are difficult to soothe. As a result, compared to other babies, they have more frequent and longer emotional episodes that tax both their developing neurobiological systems and their relationships with family members (Gunnar & Quevedo, Reference Gunnar and Quevedo2007). Moreover, these infants may be particularly dependent on the quality of caregiving to develop a broad and flexible repertoire of effective self-regulatory strategies. There is some evidence that caregiving quality moderates the relation between infant temperament and later symptoms (Bates & Pettit, Reference Bates, Pettit, Grusec and Hastings2007), which may be explained by the development of adequate skill at self-regulation of emotion. Support for the view that parenting these children is nonetheless difficult comes from evidence that when parents perceive their children as emotionally negative, the parents report more anger, more of a sense of responsibility for child misbehavior, and more use of physical punishment (Graham, Weiner, Cobb, & Henderson, Reference Graham, Weiner, Cobb and Henderson2001); however, when child characteristics are assessed objectively, child factors such as age, gender, temperament, hyperactivity, or conduct problems do not increase the likelihood of child maltreatment (National Research Council, 1993). Nonetheless, temperamental characteristics likely influence which situations young children seek and avoid, the child-rearing challenges they present to their caregivers, what emotions their parents express, and other important aspects of child–caregiver interactions, each of which has implications for child emotion socialization and the development of emotion regulation (Eisenberg, Cumberland, & Spinrad, Reference Eisenberg, Cumberland and Spinrad1998; Morris, Silk, Steinberg, Myers, & Robinson, Reference Morris, Silk, Steinberg, Myers and Robinson2007). In sum, the emotionally difficult youngster has increased risk for a variety of negative outcomes and may have parents who share dispositional qualities that compromise parenting. These parents are likely to have greater difficulty regulating their own emotions and providing optimal caregiving for the child. In turn, the difficult child is probably more dependent on the quality of parenting than are children who are less emotionally reactive.
There is, however, no direct evidence yet that the reactive, poorly regulated infant is at specific risk for the development of emotional instability and poor emotional awareness. The view of BPD as a product primarily of genetic disposition (Cloninger et al., Reference Cloninger, Svrakic and Pryzbeck1993) goes beyond the typical three factor (Rothbart & Bates, Reference Rothbart, Bates, Eisenberg, Damon and Lerner2006) and five factor (Costa & McCrae, Reference Costa and McCrae1992) models of temperament and personality; it conceptualizes personality as involving four genetically predisposed dimensions (temperament) and three acquired dimensions (character). Using the Temperament and Character Inventory (TCI; Cloninger, Przybeck, Svrakic, & Wetzel, Reference Cloninger, Przybeck, Svrakic and Wetzel1994), it was found that the combination of high harm avoidance and high novelty seeking is unique to BPD (Ferraz et al., Reference Ferraz, Vállez, Navarro, Gelabert, Martín-Santos and Subirà2009; Joyce et al., Reference Joyce, Mulder, Luty, McKenzie, Sullivan and Cloninger2003; Svrakic et al., Reference Svrakic, Whitehead, Przybeck and Cloninger1993). Harm avoidance is defined similarly to child fear proneness (avoiding and withdrawing from unfamiliar situations) and novelty seeking is defined by child surgency or high approach (approaching even unfamiliar situations with boldness or exuberance). Evidence that the TCI outperforms the five factor model in predicting borderline personality is mixed. Yet adults with BPD appear to describe themselves as having these two contradictory motivational tendencies. When applied to preschoolers, it was found that those who were high in both harm avoidance and novelty seeking also had the most serious behavior problems, both externalizing and internalizing symptoms (Constantino, Cloninger, Clarke, Bashemi, & Pryzbeck, Reference Constantino, Cloninger, Clarke, Hashemi and Przybeck2002). This is noteworthy as the co-occurrence of high internalizing and externalizing symptoms was also found among adolescents in day treatment who later developed BPD (Zelkowitz et al., Reference Zelkowitz, Paris, Guzder, Feldman, Roy and Rosval2007). The missing link, however, is evidence of the specific developmental processes that were compromised for either of these groups of children, including assessment of emotional functioning in challenging interpersonal contexts. Such research should advance our ability to draw distinctions and gain specificity about relations between early temperament and pathways to the development of different disorders.
In addition to the potential contributions of temperamental factors, it is also important to consider developmental timing and how it affects the contexts to which children are sensitive and the ways that children cope with them. Exposure to traumatic relationships has more deleterious effects on younger than older children (Manly, Kim, Rogosch, & Cicchetti, Reference Manly, Kim, Rogosch and Cicchetti2001). Environmental stress in early stages of development can also have profound effects on neurocognitive development (e.g., Milham & Banich, Reference Milham and Banich2005). A toddler or preschool age child who is maltreated by a caregiver has few options for coping, particularly if the child has already formed an insecure attachment; this may lead to a greater proclivity to rely on dissociation to cope with the complex and overwhelming emotions the child experiences. Attention control is widely viewed as an important, adaptive emotion regulation strategy in early childhood but research rarely examines at risk youngsters and their risk for dissociation or emotional avoidance (but see Cromer, Stevens, de Prince, & Pears, Reference Cromer, Stevens, DePrince and Pears2006). Early environmental stress affects attentional development (Milham & Banich, Reference Milham and Banich2005; see also Grosjean & Tsai, Reference Grosjean and Tsai2007) and maltreatment has negative effects on attentional processes (Rogosch & Cicchetti, Reference Rogosch and Cicchetti2005). Childhood trauma appears to be insufficient in predicting attention related symptoms such as dissociation (Briere, Reference Briere2006), but being young when exposed to family adversity predicts adult dissociation (Lyons-Ruth, Dutra, Schuder, & Bianchi, Reference Lyons-Ruth, Dutra, Schuder and Bianchi2006).
There is a clear need to understand characteristics of the environment that interface with the vulnerabilities of temperamentally sensitive youngsters, that may override the capacity of any child's developing emotion regulation skills, and that constitute the proximal features of at risk family environments. We now turn to a discussion of environmental risk factors, with an eye toward elements that may be most proximal in the experience of the young child.
Emotional environment as a risk factor
Linehan's model (Crowell et al., Reference Crowell, Beauchaine and Linehan2009) articulates three types of psychosocial risk for BPD that we use to conceptualize the young child's emotional environment. In referring to this environment, we are particularly interested in its palpable emotional features, that is, the elements that a young child perceives and arguably have direct effects on the child's emotional functioning. These are proximal mechanisms by which more distal risk factors compromise a young child's ability to regulate emotion and that foster emotional dysregulation. We consider the proximal emotional environment that may contribute specifically to emotional instability and poor emotional awareness.
The first psychosocial risk factor discussed by Crowell and colleagues (2009) is family history of psychopathology. They noted a high degree of impulse control disorders, and to a lesser degree mood disorders, in the families of individuals with BPD, in support of the view of shared genetic vulnerabilities to emotional dysregulation and impulsivity. Apart from studies of maternal depression, relatively little is known about how parental psychopathology affects parental emotion and how these emotional variations influence the child's development. Yet, an activity as evocative as parenting must have implications for the emotional experience, presentation, and behavior of parents and their children (Dix, Reference Dix1991). This likelihood bears directly on the second set of psychosocial risks identified by Crowell and colleagues: problematic parent–child interaction and relationships (emotional invalidation, attachment disturbance, and maltreatment). For instance, maltreatment is commonly believed to be a condition that leads to BPD because many individuals with the disorder describe abusive caregiving in their childhood (Herman et al., Reference Herman, Russell and Trocki1986; Westen, Ludolph, Misle, Ruffins, & Block, Reference Westen, Ludolph, Misle, Ruffins and Block1990). Yet most children who were maltreated do not develop the disorder, and there is substantial diversity in the outcomes of individuals who have been physically or sexually abused within the family (New et al., Reference New, Triebwasser and Charney2008). However, maltreating caregivers must vary emotionally. A closer analysis of these variations may be one way to study variations in the pathways to BPD. There may be some variants of maltreating caregiving that are more directly relevant to emotional instability and poor emotional awareness. The third psychosocial risk factor identified by Crowell and colleagues is sociocultural risk (e.g., income, cultural heritage). As they note, these environmental influences remain relatively unexplored in research on BPD, but arguably provide important contextual information. In terms of our focus on the emotional environment, cultural background, for example, likely affects the degree to which parents’ emotions and practices are construed as unloving or abusive (Cole & Tan, Reference Cole, Tan, Grusec and Hastings2006).
These three categories of psychosocial risk, of course, are not mutually exclusive. For conditions such as income or parental psychopathology to be environmental influences on the development of BPD, they must ultimately impinge on the child's immediate environment and be palpable to the child. For the very young child who may be at greater risk for the development of BPD than the older child, the emotional environment is dominated by caregiver–child relationships and interactions. Thus, the search for proximal mechanisms should, we believe, focus on the emotional environment of the family's relationships and interactions. In addition to the shared biological vulnerabilities implied by genetic risk, the immediate interactions of children with their caregivers may be the mechanisms by which the child's emotional vulnerabilities and inadequate caregiving contribute to the development of emotional dysregulation. Just as there is growing evidence that maternal depression has an effect on emotion regulation skill in offspring (Blandon, Calkins, Keane, & O'Brien, Reference Blandon, Calkins, Keane and O'Brien2008; Silk, Shaw, Skuban, Oland, & Kovacs, Reference Silk, Shaw, Skuban, Oland and Kovacs2006), there may be specific ways that the emotional environment contributes to emotional instability and poor emotional awareness.
Linehan's (1993) view of the emotional environment of the at-risk child asserts that emotionally reactive children elicit frequent high-intensity emotional exchanges with their caregivers. The caregivers respond to the child by dismissing or minimizing (invalidating) the child's emotional message as well as alternating between responding to the child with emotional intensity (which may punish the child's emotional reaction, also invalidating it) and reinforcing the child's behavior as a way of terminating the aversive emotionality. In these interactions, caregivers engage in little teaching of adaptive strategies for recognizing and regulating emotion. Thus, the reactive child is intermittently reinforced for intense emotional reactions, and does not learn to reflect on feelings (to be emotionally aware) or to manage strong emotions (to regulate them). This model intentionally draws from Patterson's (Reference Patterson1982) social learning model for the development of conduct disorder (Crowell et al., Reference Crowell, Beauchaine and Linehan2009). Despite the popularity of Linehan's model for the treatment of BPD, its hypothesis about the origins of emotional dysregulation in BPD have not been tested. The hypothesis leads to interesting research questions that could address overlap and divergence in developmental pathways to antisocial and borderline personality. A salient distinction between the emotional concomitants of these two forms of psychopathology involves interpersonal relatedness; although persons with these disorders sometimes behave similarly in interpersonal contexts, the person with antisocial personality is more likely to have reduced concern about interpersonal stress, whereas the person with borderline personality is purportedly more likely to be hurt by perceived rejection.
The attachment perspective also postulates features of the young child's emotional environment that offer specific predictions about early risk for the development of BPD (Fonagy et al., Reference Fonagy, Gergely, Jurist and Target2002; Gunderson & Lyons-Ruth, Reference Gunderson and Lyons-Ruth2008). Patterns of caregiver–child interaction lead to the development of either ambivalent or disorganized forms of insecure attachment, both of which have a higher prevalence among individuals with BPD (Agrawal, Gunderson, Holmes, & Lyons-Ruth, Reference Agrawal, Gunderson, Holmes and Lyons-Ruth2004). These forms of insecure attachment differ from secure attachment in both their origins and outcomes. Secure attachment is a product of an emotionally responsive environment, which promotes a child's confidence that distress can be relieved and that one can safely risk venturing into other relationships. A secure youngster might be emotionally intense but has come to expect that those relationships can be trusted to help the child regain emotional stability. Moreover, the sensitive caregiver will respond in a manner that implicitly, and often explicitly, affirms the child's emotional experience and promotes the young child's adaptive regulation of intense negative emotion (e.g., Gilliom, Shaw, Beck, Schonberg, & Lukon, Reference Gilliom, Shaw, Beck, Schonberg and Lukon2002; Smith, Calkins, & Keane, Reference Smith, Calkins and Keane2006).
The emotional interactions that lead to secure attachment are less likely if a caregiver's own ability to regulate emotion is impaired by conditions such as psychopathology or insecure adult attachment; this caregiver is more likely to be distressed and defensive toward the child, or to be hostile and punitive, invalidating the child's emotional experience, modeling poor emotion regulation, and failing to guide the development of child self-regulatory strategies, each and all of which could contribute to impoverished emotional awareness and emotional instability (Fonagy et al., Reference Fonagy, Gergely, Jurist and Target2002; Linehan, Reference Linehan1993). Gunderson and Lyons-Ruth (Reference Gunderson and Lyons-Ruth2008) summarized the portrait of a caregiver who is physically or emotionally unavailable, who fosters ambivalent attachment, and who frightens and/or appears frightened by the child, a pattern of caregiving that fosters disorganized attachment. These suggest palpable aspects of the young child's emotional environment. Research on parental emotion, however, has been slow to emerge despite Dix's (1991) call for its study. Recent evidence indicates that mothers with insecure adult attachment are less accurate reading their infants’ anger and fear and are more likely to react to infant distress with affective neutrality or amusement (Leerkes & Siepak, Reference Leerkes and Siepak2006). Such responses likely confuse a very young child, invalidate the child's experience, and contribute to the development of poor emotional awareness through insensitivity and inaccuracy and perhaps to infant emotional instability by being insensitive and perhaps unpredictable. Infant distress, particularly among those who are reactive and hard to soothe, may escalate when mothers are this insensitive; infant temperament, unchecked by parental succorance and guidance, contributes to the maintenance of child emotional intensity, and therefore emotional instability over time (e.g. Crowell et al., Reference Crowell, Beauchaine and Linehan2009).
Potential influences of parental emotion on children's emotional development can also be drawn from maltreatment studies. First, the research demonstrates that young maltreated children appear less aware of their own and others’ emotions than nonmaltreated children (Beeghly & Cicchetti, Reference Beeghly and Cicchetti1997; Camras, Grow, & Ribordy, Reference Camras, Grow and Ribordy1983; Camras, Sachs-Alter, & Ribordy, Reference Camras, Sachs-Alter, Ribordy, Lewis and Sullivan1996; Pears & Fisher, Reference Pears and Fisher2005; Rogosch, Cicchetti, & Aber, Reference Rogosch, Cicchetti and Aber1995), which appears to mediate the relation between maltreatment and emotion dysregulation (Izard et al., Reference Izard, King, Trentacosta, Morgan, Laurenceau and Krauthamer-Ewing2008; Rogosch et al., Reference Rogosch, Cicchetti and Aber1995). Second, although this poorer emotion awareness seems to involve several emotions, anger appears to be a special case (Fries & Pollak, Reference Fries and Pollak2004; Pollak, Cicchetti, Hornung & Reed, Reference Pollak, Cicchetti, Hornung and Reed2000), notable because maltreating caregivers may engage in more child-directed anger (Pollak, Reference Pollak, Cicchetti and Gunnar2009). This may explain why physically abused children more easily recognize anger but not sadness and disgust (Pollak et al., Reference Pollak, Cicchetti, Hornung and Reed2000), and why the speed with which children detect anger in a face contributes to relations between physical abuse and child anxiety (Shackman, Shackman, & Pollak, Reference Shackman, Shackman and Pollak2007). That is, in addition to a greater likelihood of being insecurely attached (Cicchetti, Toth, & Bush, Reference Cicchetti, Toth, Bush, Lahey and Kazdin1988), being exposed to child-directed anger by the person who is supposed to protect the child appears to have significant impact on how children anticipate, understand, and react to anger. This work directly addresses the proximal emotional environment by examining at risk children's processing of nonverbal anger cues.
In addition to exposing children to parental emotions that children may find difficult to understand, the abusive emotional environment probably is emotionally overwhelming while providing fewer opportunities to learn to regulate emotion in healthy ways. In thinking about children's ability to cope with parental emotions that overwhelm them, it is important to not focus solely on child emotional reactivity or intensity. Children in these contexts, although they may still long for closeness with the caregiver, may stop expecting emotional support. In this instance, some children may learn to hide or even avoid their emotions, a pattern that can become a stable means of coping. Maltreated children indicate reluctance to express emotions to their parents (Shipman, Zeman, Fitzgerald, & Swisher, Reference Shipman, Zeman, Fitzgerald and Swisher2003) and can respond quickly and compliantly to parental requests (Crittenden & DiLalla, Reference Crittenden and DiLalla1988). Children exposed to other stressful family contexts, such as interparental conflict or parental disorganized attachment, tend to reverse roles with their caregivers (Macfie, Fitzpatrick, Rivas, & Cox, Reference Macfie, Fitzpatrick, Rivas and Cox2008; Macfie, Houts, Pressel & Cox, Reference Macfie, Houts, Pressel and Cox2008). Although maltreated children are observed to be emotionally negative, reactive, and labile in many contexts (Berliner & Elliott, Reference Berliner, Elliott, Myers, Berliner, Hendrix, Terry and Jenny2002; Maughan & Cicchetti, Reference Maughan and Cicchetti2002; Parker & Herrera, Reference Parker and Herrera1996; Pollak & Sinha, Reference Pollak and Sinha2002; Shields & Cicchetti, Reference Shields and Cicchetti1998; Teisl & Cicchetti, Reference Teisl and Cicchetti2008), which may be mediated by impoverished maternal support of child emotion (Shipman & Zeman, Reference Shipman and Zeman2001), which is observed in maltreated children (Maughan & Cicchetti, Reference Maughan and Cicchetti2002; Pollak & Sinha, Reference Pollak and Sinha2002; Shields & Cicchetti, Reference Shields and Cicchetti1998; Teisl & Cicchetti, Reference Teisl and Cicchetti2008), it is important to appreciate the evidence of oversolicitousness and emotional shutdown that can also be seen. What is not known is whether child reluctance to reveal emotion to parents coupled with a proclivity to be emotionally reactive leads to the emotional instability typified by BPD. Although some maltreating families may be characterized by escalating cycles of coercive interaction as Linehan (Reference Linehan1993) suggested, there is evidence to suggest that maltreating families communicate rigidly with little emotion expression (Cicchetti & Howes, Reference Cicchetti and Howes1991). Analysis of the proximal emotional environment may aid our understanding of commonalities among effects of child maltreatment and other risk factors, such as marital conflict and parental disorganized attachment. Children are not passive bystanders to adult emotions; at least one study indicates that children perceive themselves as agents in the face of parental difficulties (Schermerhorn, Cummings, & Davies, Reference Schermerhorn, Cummings and Davies2005). Therefore, a child's exposure to adult anger, directed at the child or to others, may be met with child fear and avoidance, or concern and effort to provide care, and these may have different consequences for the patterns of emotional regulation and dysregulation that children develop. Having summarized the evidence of the effects on emotional functioning in early childhood of the child and family risk factors purported to contribute to BPD, we next consider the developmental trajectories that might link early childhood emotional difficulties with the emotional instability and poor emotional awareness of adults with BPD.
Developmental paths to borderline emotional instability and poor emotional awareness
Emotional instability and poor emotional awareness are cardinal features of the emotional dysregulation associated with BPD. Theory and evidence characterize risk for the origins of BPD as an interplay of child characteristics, particu- larly temperamental reactivity or sensitivity, and family factors, largely significant deviations in the quality of parent–child interactions attributable to both biological and behavioral risk (e.g., shared genetic risk, child maltreatment). Furthermore, the relation between these risk factors and the development of BPD may depend on the age of the at risk child. The earliest periods of life, including the prenatal period, may carry the greatest risk, because of both the vulnerability of developing neurobehavioral systems that support emotion regulation and the limited options for very young children for coping with strong reactions to emotionally challenging environments.
Evidence of the effects of putative risk factors on child emotional functioning has not, at least as yet, indicated homotypic continuity with the forms of emotional dysregulation that typify borderline personality, although most theories assert that the environmental and genetic roots of the disabling symptoms of BPD exert their influence in the first years of life. To investigate the nature of the continuities between early childhood risk and the development of chronic emotional instability and poor emotional awareness, we advocate a two-pronged approach. First, we believe that a clinically informed assessment of the biological and behavioral aspects of emotion regulation will heighten the opportunity to capture more finely grained, pertinent aspects of emotional functioning (Cole & Hall, Reference Cole, Hall, Beauchaine and Hinshaw2008). That is, it is important to move beyond examination of general emotional negativity or reactivity and to investigate multiple, different emotions (e.g., anger followed by shame or anxiety followed by emotional avoidance), the temporal dynamics of emotional responses (including the latency and intensity of responses; Thompson, Reference Thompson1994), the context appropriateness of emotional responses such that exaggerated or inappropriate reactions can be distinguished from strong but understandable reactions, and the relation between emotional reactions and their behavioral consequences, particularly behaviors that involve cutting off or harming the self or others (Cole & Hall, Reference Cole, Hall, Beauchaine and Hinshaw2008). Second, we advocate the thorough and careful examination of the young child's proximal emotional environment. Such an analysis of the environments that pose risk should aid specifying the nature of these risks, including explaining why different risk factors have common outcomes and single risk factors lead to different outcomes. That is, if we focus on the emotional environment of the young child, and the emotional dynamics of each child's particular sensitivities to its features, we may articulate the development of chronic emotional instability and poor emotional awareness.
Specifically, capturing the emotional environment that fosters borderline emotional dysregulation requires examination of the features of the interpersonal surround that are perceptually salient to the young child, individual differences in child sensitivity to these features, and the effects of those features on the child's neurobehavioral functioning, with special attention to emotion, attention, and social outcomes. This approach therefore requires experimental designs that identify emotional features that children perceive such as nonverbal emotional information (Pollak, 2009) and longitudinal designs that examine the implications of these individual differences for child functioning over time. We underscore the importance of studying emotion in context, particularly in interpersonal context. Capturing the development of emotional instability and poor emotional awareness will require assessment of multiple levels of analysis, including individual, dyadic, and family level biological and behavioral functioning and macro-level factors that impinge on these.
Emotional instability
It is generally assumed that young children have intense emotional reactions that can change quickly, but there is little evidence to address this assumption. There are a few studies of emotional instability in older children, but they do not distinguish emotional reactivity from instability (DeClercq, DeFruyt, Van Leeuwen, & Mervielde, Reference DeClercq, DeFruyt, Van Leeuwen and Mervielde2006; Leech, Day, Richardson, & Goldschmidt, Reference Leech, Day, Richardson and Goldschmidt2003). An exciting new direction then is the study of the development of emotional instability. Such work should distinguish the intensity of an emotional reaction (intensity and latency to peak intensity) from the nature of emotional shifts (intense positive emotion followed by negative emotion), relating these to situational context (e.g., fear in reaction to a stranger is different than fear in reaction to a parent), and addressing their predictability and appropriateness. Again, this also requires sensitivity to the fact that multiple emotions are involved when a parent overwhelms a young child emotionally. Fear, anger, and shame, to name a few, are logical responses to disturbances in close relationships.
In the adult literature, emotional instability (e.g., Ebner-Priemer et al., Reference Ebner-Priemer, Kuo, Kleindeinst, Welch, Reisch and Reinhard2007; Trull et al., Reference Trull, Solhan, Tragesser, Jahng, Wood and Piasecki2008) has been studied in ways that can inform the study of risk for borderline emotional dysfunction in children, using time series data and multilevel assessment (context, emotion-related behavior, and neurobiological processes). This work emphasizes the importance of studying emotional changes in temporal context (e.g., emotional shifts) rather than as variability around a mean level of emotion. Emotional shifts are rarely studied in young at risk children but one study offers a glimpse of the fruitfulness of assessing them. During a self-recognition test, toddlers of depressed mothers were more likely to shift from positive to nonpositive emotion upon recognizing themselves in a mirror, a pattern that was not typical of toddlers of nondepressed mothers (Cicchetti, Rogosch, Toth, & Spagnola, Reference Cicchetti, Rogosch, Toth and Spagnola1997). The findings highlight the importance of observing emotion in context. In sum, future research should investigate both typical and at risk patterns of emotion changes, including multiple emotions, linking them to context and observing their implications over time.
Poor emotional awareness
Emotion cognition has received attention in the child development literature, but not necessarily in ways that address the poor emotional awareness that typifies BPD. Nonetheless, there is ample evidence that very young children discern and use emotional information, particularly in terms of facial expressions and affective prosody (Fernald, Reference Fernald1993; Klinnert, Emde, Butterfield, & Campos, Reference Klinnert, Emde, Butterfield and Campos1986; Repacholi & Meltzoff, Reference Repacholi and Meltzoff2007). By ages 3 through 5 when children demonstrate accuracy in emotion knowledge (Denham, Reference Denham1998; Harris, Reference Harris1989), they have had ample experience with their emotional environments. Young children who are at risk by virtue of one or more purported BPD risk factors show atypicalities in emotion cognition. Maltreated children who are exposed to significant parental negative emotion label emotions inaccurately, including attributing anger to nonangry expressions (Camras et al., Reference Camras, Grow and Ribordy1983; Pollak et al., Reference Pollak, Cicchetti, Hornung and Reed2000). They are also quick to detect angry expressions and threat-related cues (Pollak et al., Reference Pollak, Cicchetti, Hornung and Reed2000; Shackman et al., Reference Shackman, Shackman and Pollak2007) and slow to shift attention back to a location where they saw angry expressions (Pine et al., Reference Pine, Mogg, Bradley, Montgomery, Monk and McClure2005). The findings underscore the need to examine both emotional and cognitive functioning when children perceive threatening environmental cues. For example, a history of maltreatment and poor attention control predicted potential precursors of BPD (emotional lability, low conscientiousness, self-rated and other-rated relationship problems, self-harm) in maltreated children (Rogosch & Cicchetti, Reference Rogosch and Cicchetti2005).
Research on attentional processing in anxiety may help to shed light on this pattern of findings. Anxious individuals are vigilant to threat-related material (e.g., MacLeod, Mathews, & Tata, Reference MacLeod, Mathews and Tata1986; Mathews, & MacLeod, Reference Mathews and MacLeod1986), but their memory of the information is inhibited (Mogg, Mathews, & Weinman, Reference Mogg, Mathews and Weinman1987), suggesting a vigilance-avoidance pattern in response to threat. This same pattern may occur for young children exposed regularly but perhaps unpredictably to threat in their families. The data suggest that exposure to hostility and anger in the emotional environment fosters child insecurity and anxiety, heightens a child's ability to detect anger, and leads to avoidance of threatening stimuli. Hypervigilance to threat-related cues paired with subsequent avoidance of those cues (Herman, Reference Herman1992; Hennessy, Rabideau, Cicchetti, & Cummings, Reference Hennessy, Rabideau, Cicchetti and Cummings1994; Pollak et al., Reference Pollak, Cicchetti, Hornung and Reed2000; Rieder & Cicchetti, Reference Rieder and Cicchetti1989), particularly in family contexts that dismiss or discourage conversing about emotional experiences, may lead to poor emotional awareness, which could exacerbate emotional instability. A unique pattern of vigilant scanning for environmental dangers, but avoiding, denying, or dissociating those experiences has not been studied comprehensively in children at risk for BPD.
Temperamental contributions
The juxtaposition of high harm avoidance and high novelty seeking in adults with BPD offers interesting hypotheses for developmental study. These two aspects of temperament do not tend to occur in most children, and there are no studies that we could find that considered them in concert. Yet both are associated with higher cortisol activity, an index of stress, in young children (e.g., Donzella, Gunnar, Krueger, & Alwin, Reference Donzella, Gunnar, Krueger and Alwin2000), although the link between stress and these emotional dispositions may depend on low effortful control. Two possibilities emerge. One is that effortful control is the key factor in BPD because it involves the ability of the child to regulate both fearful (harm avoidance) and surgent (novelty seeking) tendencies (Crowell et al., Reference Crowell, Beauchaine and Linehan2009; Grosjean & Tsai, Reference Grosjean and Tsai2007). The other is that emotional environments that foster BPD may interfere with the development of effortful control or executive function (e.g., Fonagy et al., Reference Fonagy, Gergely, Jurist and Target2002), perhaps because these contexts are both highly evocative and fail to support learning to reflect on and integrate experience.
The proximal emotional environment
A focus on the emotional nature of the environment may yield common links among parents who are stressed, have psychopathology, have conflict with each other, are abusive, and/or minimize and dismiss children's emotions. That is, each of these factors may be associated with an emotional environment that (a) emotionally overwhelms a child, (b) fails to provide adequate strategies for regulating intense emotion, (c) may reinforce child emotional intensity but also emotional avoidance, and (d) is unpredictable and places the child in a chronic state of uncertainty or insecurity. Witnessing one parent assault the other, having a parent engage in inappropriate sexual activity, interacting with a parent who responds to child distress with bemusement or anger: each of these is likely to elicit strong emotions in a child, all the more so an emotionally reactive child who is low in dispositional effortful control. Note that we say strong emotions, in the plural; again it is important to appreciate the multiple emotions that these situations can engender: fear, anger, disgust, guilt or shame, and in some cases pleasurable emotions also, that contribute to being emotionally overwhelmed and may become patterns of response that acquire maladaptive qualities.
If we consider the period from toddler to preschool age when children begin to engage in purposive self-regulation of distress and negative emotions (Kopp, Reference Kopp1989), it is easy to imagine how easily these risk-ridden emotional environments overwhelm a young child's ability to cope. Moreover, for many individuals with BPD the historical root of the emotions that overwhelm is a caregiver, the person with whom the child longs to be related. A corollary is that such caregivers either fail or are inconsistent in their behavior that models emotion, discusses emotional experience, and teaches emotion regulation strategies, particularly in situations in which they overwhelm the child. Young children may be emotionally reactive in these contexts but some are not when threatened by a caregiver; some may shut down, possibly engaging in forms of emotional avoidance that contribute to emotional instability and poor emotional awareness. There is ample evidence that children from high conflict families are stressed, as indexed by both neurophysiological and behavioral measures. A promising new direction then is the assessment of the specific features of the environment that stress and overwhelm these systems.
Conclusions
In sum, emotional instability and poor emotional awareness are cardinal features of the emotional dysregulation associated with BPD, and these regulation failures likely have their roots in early periods of development. There is a need to understand the development of these aspects of emotional functioning. A few studies indicate that children who are at risk for developing BPD may have high levels of both internalizing and externalizing symptoms in childhood or adolescence. The links between this pattern of high symptom levels and specific patterns of at risk emotional functioning need to be determined. Developmental work on various aspects of emotion cognition provide a modest normative base from which to study the development of typical and impoverished emotional awareness, but this work has not been fully applied to the study of children at risk for BPD. A valuable direction is to appreciate the complex nature of emotional sensitivity, awareness, and avoidance and how it might develop in emotional environments that overwhelm and confuse young children, particularly those who may be emotionally reactive to environmental sensitization and/or by genetic predisposition. There is less of an empirical base from which to study the developmental course of emotional stability and how the purported risk factors affect it. Work with adult emotional instability affords some suggestions for research methods, with methods that lend themselves to examining emotions in temporal context and linked in time at both physiological and observational levels of analysis with young children and including experiential levels with older children.
Emotional instability and poor emotional awareness are predicted to emerge as products of a child's heightened reactivity, whether that reactivity is driven by biological predisposition or sensitivities that result from exposure to stress, pre- and/or postnatally, and emotional environments that overwhelm the child without helping the child learn to cope with the overwhelming emotions. Factors such as child age at the onset of emotionally overwhelming family circumstances and easy temperament, and family factors such as the availability of one or more emotionally responsive adults, may help us to understand individual differences in how children react to and cope with parental emotions that overwhelm them. Because the origins of BPD are attributed to early childhood, the study of at-risk emotional environments, with an emphasis on the features of the environment that are perceived by very young children, including facial expressions and vocal quality, should yield new and important information. The youngest children detect and differentiate emotional variations in those around them, even before they can assign significant meaning to them. We advocate for detailed examination of emotional responses, at multiple levels of analysis, and focusing on multiple rather than single emotions as unfolding processes that occur in response to specific features of children's emotional environments. Examining these proximal environments, children's individual sensitivities to them, and the nuanced nature of their emotional responses should reveal more about how chronic emotional instability and poor emotional awareness develop. Furthermore, this approach may help to understand why a variety of risk factors can lead to BPD and yet each of those risk factors is also associated with diverse outcomes, including healthy outcomes. Knowing the environmental factors that have sensitized a child to emotional dysregulation should improve the methods we use to identify BPD risk, to intervene with at risk children, and to prevent the development of such a disabling form of psychopathology.
