Extreme levels of childhood adversity have been consistently associated with lifelong and transdiagnostic risk of psychopathology (Green et al., 2010; Kessler et al., 2010). Recent efforts have focused on understanding the underlying psychological mechanisms, and progress in this area is crucial for increasing the specificity of psychological interventions (Dvir, Ford, Hill, & Frazier, 2014; Nanni, Uher, & Danese, 2012). Emotion regulation stands out among candidate mechanisms in light of its association with both childhood adversity (for review see Dvir et al., 2014; Hoppen & Chalder, 2018; Jaffee, 2017; McCrory, De Brito, & Viding, 2012; McLaughlin, 2016), and psychopathology (Aldao, Gee, De Los Reyes, & Seager, 2016; Beauchaine & Cicchetti, 2019; Gross & Jazaieri, 2014; Kring, 2008). However, relatively few studies have directly examined the mediator role of emotion regulation, and the integration of these results is not straightforward given the multiplicity of theoretical approaches to emotion regulation. We capitalized on recent advances in meta-analytic structural equation modeling to investigate the mechanistic involvement of emotion regulation in the relation between childhood adversity and psychopathology. To do so, we used extensive data on relevant pathways (i.e., childhood adversity to emotion regulation, and emotion regulation to psychopathology), as available in published studies.
Almost all forms of psychopathology involve a component of disrupted emotion regulation, making this concept a candidate mechanism in which to explore individual differences in the psychological consequences of stressful events (e.g., Lazarus & Folkman, 1984). In what is probably the first study on childhood adversity and coping, Leitenberg, Greenwald, and Cado (1992) examined the habitual use of multiple strategies in a sample of women who reported a history of childhood sexual abuse. Among the strategies assessed were rumination, cognitive reappraisal, and expressive suppression, and results indicated that, in this sample, suppression was used more frequently than other strategies, and it was negatively associated with psychological adjustment after controlling for multiple characteristics (e.g., age when abuse began, duration) of sexual abuse (Leitenberg et al., 1992). Another early study used a broader assessment “targeting processes central to emotionality and regulation, including affective lability, intensity, valence, flexibility, and situational appropriateness of emotional expressions” (Shields & Cicchetti, 1998 p. 385) and found that maltreated children (identified based on official social services records) showed higher emotional lability and negativity, and lower adaptive emotion regulation, as well as higher aggressive behavior compared to non-maltreated children. The effect of maltreatment on aggression became non-significant when controlling for emotional lability/negativity, which suggested that the latter may play a mediator role (Shields & Cicchetti, 1998). These studies have laid the groundwork for future investigations, which would incorporate other emotion regulation conceptualizations (e.g., Gratz & Roemer, 2004; Gross, 1998; Nolen-Hoeksema, 1991) that emerged in clinical psychology. Studies on the development of emotion regulation have also suggested that this domain is likely to be vulnerable to childhood adversity. From a developmental perspective, emotion is a property that emerges in a cascading manner from abilities including attention, memory, theory of mind, and categorization—each of which may be influenced by childhood adversity (Ruba & Pollak, 2020). Similarly, emotion regulation subsumes many aspects and processes which undergo developmental changes throughout childhood and adolescence. For instance, emotion understanding improves in early childhood, the repertoire of emotion regulation strategies expands during adolescence, and the habitual use of certain emotion regulation strategies, particularly cognitive strategies (e.g., reappraisal), which allow individuals to regulate emotion while remaining engaged in the processing of emotional events, becomes consistent in adolescence and emerging adulthood (for review see Compas et al., 2017; R. A. Thompson & Goodman, 2010). These developmental changes are thought to be driven by an interplay between the maturation of neuroendocrine systems and social learning opportunities, both of which may be particularly sensitive to childhood adversity. On the one hand, the immature status of the hypothalamic-pituitary-adrenal axis and the prefrontal-amygdala circuits, for example, renders these stress systems susceptible to lifelong structural and functional alteration following childhood adversity (for review see Lupien, McEwen, Gunnar, & Heim, 2009; McCrory et al., 2012). On the other hand, youth learn how to regulate their emotions mostly from their parents, through modeling and feedback (Kiel & Kalomiris, 2015; Morris, Criss, Silk, & Houltberg, 2017). Maltreating parents consistently report poor emotion regulation (Lavi et al., 2019; Lavi, Ozer, Katz, & Gross, 2021; Plate et al., 2019) and their influence may bias social learning of emotion regulation in children. Recent studies support this view: not only that parental emotion regulation difficulties are associated with children's emotion regulation difficulties (Osborne, Duprey, O'Brien Caughy, & Oshri, 2021), but the former may fully account for the impact of childhood adversity on children's emotion regulation (Milojevich, Machlin, & Sheridan, 2020). In summary, childhood adversity may negatively interfere with emotion regulation development through both premature activation of biological stress systems and maladaptive social learning.
One challenge to meta-analysis is related to the multiplicity of approaches to emotion regulation, which precludes the combination of all emotion regulation measures in a single global outcome. As illustrated in early studies (Leitenberg et al., 1992; Shields & Cicchetti, 1998), emotion regulation has been conceptualized along two main lines, which have developed relatively independently in different fields of psychology and are distinct at several levels (for review see John & Eng, 2014). One focuses on the functional effects of emotion regulation, as reflected by cognitive performance and goal-directed behavior in the presence of emotion (Cole, Michel, & Teti, 1994; Gratz & Tull, 2010; R. A. Thompson, 1994). From this perspective, emotion regulation, whether assessed as ability or difficulty, has been defined as the capacity to keep the balance between the urgency of emotion and competing goals (Cole et al., 1994), or modulating emotional arousal so as to foster an optimal level of engagement with environment (Shields & Cicchetti, 1998). The scope of emotion regulation assessments in this category is broad and includes: (a) effects of emotion on domains such as attentional control, goal pursuit, and social expression; (b) emotional awareness and the perception of emotional characteristics such as frequency, intensity, and lability; and (c) emotional acceptance, viewed as a fundamental condition of emotion regulation (opposite emotional control or the avoidance of emotion altogether) (for review see Gratz & Tull, 2010). The second category includes process approaches to emotion regulation, focusing on strategies that are commonly employed in efforts to modulate emotional responses according to goals. These strategies (some of which have also been studied in the coping literature, see Compas et al., 2014) have been systematically characterized based on the stage of emotional processing in which they intervene (Gross, 1998, Gross, 2015), as well as their relation to psychopathology (Nolen-Hoeksema, 1991, Nolen-Hoeksema, 2004), and assessments have mostly focused on their habitual use (Gross & John, 2003; Treynor, Gonzalez, & Nolen-Hoeksema, 2003). Overall, both approaches (i.e., difficulties/ability, and strategies) have been used in research on childhood adversity, and provide complementary information that needs to be analyzed separately.
Two recent meta-analyses have examined the relation between childhood adversity and emotion regulation. One (Lavi, Katz, Ozer, & Gross, 2019) included only studies (k = 58) in children and adolescents aged up to 18, which met one of two criteria: maltreatment was substantiated based on official records, and longitudinal approach. Outcomes related to both emotion reactivity and emotion regulation were considered, and further coded separately depending on their focus on subjective experience, behavior, or physiology. The results supported increased negative affect in maltreated compared to non-maltreated youth, both at the level of experience and behavior, as well as decreased positive affect and increased aggression at the behavioral level. In which emotion regulation was concerned, maltreated youth showed increased difficulties, as indexed by a variety of outcomes related to both emotion regulation, and coping (Lavi, Katz, et al., 2019). The other meta-analysis (Gruhn & Compas, 2020) included 35 studies in children and adolescents (age 5–18), which focused on childhood maltreatment, and emotion regulation and coping. Outcomes were categorized at three levels of generality, from broad domains (e.g., total emotion regulation/coping) to factors (e.g., emotion regulation/coping styles) and specific strategies. Childhood maltreatment was significantly associated with both domain measures, that is, negatively with emotion regulation/coping ability and positively with emotion regulation/coping difficulties. Significant associations were also found at the strategy level, including positive relations between maltreatment and emotional suppression (as well as expression), but not distraction (Gruhn & Compas, 2020). No relations were found at the factor level. These meta-analyses have lent support to the relation between childhood adversity and emotion regulation in youth. However, both left out studies in adults, which are the largest part of the empirical literature, and excluded forms of childhood adversity other than maltreatment. Most forms of psychopathology start to emerge in adolescence and adulthood, so it is not surprising that studies investigating the involvement of childhood adversity and emotion regulation in risk of psychopathology have overly focused on adults. Some of these studies have supported the negative impact of a variety of adverse childhood experiences such as extreme poverty, natural disasters, war, or parental psychopathology on emotion regulation and risk of mental disorders, suggesting that these effects are not limited to maltreatment. In addition, and perhaps most importantly, previous meta-analyses have not addressed the mediator role of emotion regulation in the relation between childhood adversity and psychopathology, which holds the most clinical relevance. The relations between emotion regulation and psychopathology have also been consistently supported in meta-analyses. For example, suppression and rumination have been positively associated (Aldao, Nolen-Hoeksema, & Schweizer, 2010; Compas et al., 2017; Hu et al., 2014; Prefit, Candea, & Szentagotai-Tatar, 2019; Seligowski, Lee, Bardeen, & Orcutt, 2015), and reappraisal has been negatively associated (Aldao et al., 2010; Hu et al., 2014; Prefit et al., 2019; but see Seligowski et al., 2015) with psychopathology. It is noteworthy that some of these relations (e.g., suppression and psychopathology) were found to be larger in adults compared to children and adolescents (Aldao et al., 2010). Broader measures of emotion regulation, such as difficulties (Prefit et al., 2019; Seligowski et al., 2015) and abilities (Compas et al., 2017), have also shown consistent associations with internalizing and externalizing symptoms. These results support the mechanistic involvement of emotion regulation in psychopathology, and further underscore the hypothesis that emotion regulation may be one of the mediators underlying the impact of childhood adversity on mental health.