Defining Adversity
In most research to date, ACEs have been operationalized using the adversities examined in the seminal ACE Study, and indexed using the ‘ACE score’ produced by summing the number of those experiences (Felitti et al.,
1998). Approaches to the definition and conceptualisation of ACEs have nonetheless evolved considerably over time, and much has been written about the limitations of relying on the ACE score for this purpose (Lacey & Minnis,
2020). It is noteworthy that no formal definition of adversity was provided in the original ACE study, and there was no theoretical basis to the selection of the specific adversities included. Moreover, subsequent research has continued to suffer from a lack of consistent definitions. There is much to be gained by the collective adoption of a definition that addresses both developmental and ecological aspects of adversity, such as that proposed by McLaughlin (
2016), which defines childhood adversity as experiences that are likely to require significant adaptation by a typical child and that represent a deviation from the expectable environment (i.e., the environmental inputs that the human brain requires to develop normally).
The ACE Construct
The adversities specified in the original ACE Study relate largely to the micro- (individual) and meso-(family) system levels of influence articulated in ecological models of human development (e.g., Bronfenbrenner,
1986). These adversities span three broad categories of abuse (emotional, physical, sexual), neglect (emotional, physical), and household dysfunction (caregiver mental illness, substance use, incarceration, domestic violence, separation or divorce) (Felitti et al.,
1998). Given that some experiences of adversities such as divorce and parental psychopathology will not meet the definition of adversity proposed, McLaughlin (
2016) argued that they should only be indexed as ACEs when resulting in significant disruptions to caregiving or other significant adversity (e.g., consistent unavailability and unresponsiveness; emotional abuse).
Research based on an expanded approach to operationalising the ACEs construct has incorporated a greater emphasis on exo-level (extra-familial) adversities, including racial discrimination, placement in foster care, living in a disadvantaged neighborhood, witnessing violence, and bullying (e.g., Cronholm et al.,
2015). It has been further argued that crisis migration (e.g., relocation due to fleeing large-scale emergencies such as armed conflicts, natural disasters, government repression) should also be recognized as an ACE, given its potential to disrupt development in many populations worldwide (Ertanir et al., this issue). Much remains to be learned about the relative contribution of these adversities to child mental health. Folk et al. (this issue), for example, found that while the original ACEs predicted behavioral outcomes among justice-impacted youth, this prediction was not significantly enhanced by expanded ACEs. This likely reflects the more proximal effects of micro- (individual) and meso- (family) systems on developmental pathways.
Cumulative Risk and Dimensional Models
Implicit in research based on an overall ACE score is the assumption that discrete forms of adversity have additive effects on developmental outcomes, and that no single adversity is more important or influential than another. Although many studies have reported a roughly stepwise progression in risk based on the ACE score, findings have been highly heterogeneous (Hughes et al.,
2017), and it has been estimated that about 30–40% of variance in outcomes is accounted for by additive synergistic interactions between certain pairs of individual ACEs (Briggs et al.,
2021). To understand the contributions of ACEs to development and mental health it appears important to distinguish between different types of adversity, and various approaches to this can be found in the literature. Weems et al. (
2021), for example, emphasized the distinction between ACEs that meet diagnostic criteria to be classified as traumatic stressors (e.g., physical abuse; sexual abuse; witnessing violence; natural disasters) versus those that typically do not (e.g., emotional abuse; caregiver mental illness; caregiver incarceration). This recognizes the importance of how adversities are subjectively experienced, which in the context of traumatic stressors includes perceptions of extreme threat (e.g., potential death, serious injury, sexual violence; American Psychiatric Association,
2013).
Researchers informed by evolutionary life-history theory have distinguished between dimensions of adversity related to unpredictability and harshness (Ellis et al.,
2022). Unpredictability in early caregiver-child interactions (e.g., lack of contingent responsivity to infant cues), shapes children’s expectations of the environment and undermines not only the formation of secure attachment, but broader cognitive and socioemotional development (Ugarte & Hastings,
2022). Unpredictability in the child’s broader ecology, such as variability in housing, parental employment, and parental involvement in care, may further affect development by disrupting caregiving behavior. Conversely, harshness is an overarching concept encompassing experiences spanning threat and deprivation. McLaughlin et al. (
2014) further conceptualized threat and deprivation as separate dimensions. Threat experiences are those associated with actual harm or threat of harm to survival, including direct victimization (e.g., physical abuse), as well as victimisation witnessed by the child (e.g., interparental violence). Deprivation involves a lack of social and cognitive inputs from the environment that reduces opportunities for learning (e.g., neglect). As discussed further in the following sections, these models recognize that experiences of adversity are often complex and co-occurring, while attempting to distill these experiences into core underlying dimensions that cut across multiple forms of adversity and shape learning and patterns of brain development in distinct ways.
Although ACEs increase risk for many childhood disorders, these disorders often also develop in the absence of adversity. This reflects the principle of equifinality in developmental psychopathology, whereby a single outcome may originate from different risk factors (Cicchetti & Rogosch,
1996). For example, adversity may feature heavily in the risk pathway to oppositional defiant disorder in one child, whereas this same disorder may arise from largely separate risk factors in another (Hawes et al., in press). Multifinality, conversely, refers to the multiple outcomes that may result from the same risk factor, as reflected in the broad range of disordered and healthy outcomes that can follow from ACEs. It has been argued that dimensional models of ACEs are particularly key to identifying the risk mechanisms that may account for such multifinality, and likewise, informing targeted intervention and prevention strategies (McLaughlin,
2016; Weems et al.,
2021).
These conceptualisations have begun to inform novel studies of ACEs and child mental health involving a range of innovative methodological approaches. This includes research by Sisitsky et al. (this issue), in which person-centred analysis was used to identify latent profiles based on levels of threat and deprivation experienced by children at age 3 years, and found that later externalising and internalising symptoms (age 9) were best captured by a model that included four distinct types of exposure (home threat, community threat, lack of stimulation, and neglect). Similarly, in a longitudinal study of children (mean age 9 years) followed into early adulthood, latent profiles based on type of ACEs (Low Exposure, Familial Dysfunction, Emotional Maltreatment, Pervasive Exposure), were differentially associated with externalising and internalising outcomes (Nguyen et al., this issue). Research further suggests that the distinction between such dimensions may help explain intergenerational risk pathways. For example, Lyons-Ruth et al. (this issue) found that threat versus deprivation dimensions of mothers’ own ACEs were differentially associated with patterns of brain development in their infants. Specifically, a maternal history of childhood neglect, but not abuse, was associated with lower infant grey matter volume, whereas abuse, but not neglect, was associated with smaller infant amygdala volume later in development.