Child Maltreatment,Autonomic Nervous System Responsivity,and Psychopathology: Current State of the Literature and Future Directions

Abstract

Child maltreatment may affect autonomic nervous system (ANS) responsivity, and ANS responsivity may influence the impact of child maltreatment on later outcomes including long-term mental/physical health. This review systematically evaluated the evidence regarding effects of maltreatment on ANS responsivity in children and examined how ANS responsivity may influence the association between maltreatment and psychopathology, with attention to relevant developmental issues. We searched the literature for relevant studies using Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. We searched five electronic databases, performed key word searches in relevant journals, hand searched reference sections of relevant articles, and contacted experts in the field. Articles were extracted according to inclusion criteria and their quality assessed. The search produced 1,388 articles; 22 met inclusion criteria. Most of the studies suggested blunted cardiovascular responsivity generally and sympathetic activation specifically in response to stress in maltreated children compared to nonmaltreated children. Findings around vagal responsivity and skin conductance were mixed. Limited evidence was found for ANS responsivity as a moderator or mediator of psychopathology risk among maltreated children. Maltreatment may be associated with blunted sympathetic activation in stressful situations. Differences in ANS responsivity may influence psychopathology risk among maltreated children. Further research is needed to confirm the nature and magnitude of such effects.

Keywords

psychopathology maltreatment children stress autonomic nervous system

Background

Child maltreatment (physical abuse or neglect, sexual abuse, emotional abuse or neglect, witnessing domestic violence) is associated with numerous negative outcomes throughout the life span (Nemeroff, 2016), including dysregulation of the human stress response (Alink, Cicchetti, & Kim, 2012). A normal response to stress involves activation of the autonomic nervous system (ANS), followed by activation of the hypothalamic–pituitary–adrenal (HPA) axis, which produces cortisol in humans, and subsequent deactivation of the ANS. Timely activation and deactivation of the stress response allow an individual to manage threat and return to normal functioning following threat cessation. Conversely, abnormal stress responses include a prolonged “hyperresponsivity” or a diminished “hyporesponsivity” response. Both hyper- and hyporesponsivity of stress systems may result in an inability to respond adaptively to a stressor (Hunter, Minnis, & Wilson, 2011). Such stress responses may have long-term negative consequences for emotional and cognitive functioning and contribute to the development and maintenance of psychopathology (De Kloet, Joels, & Holsboer, 2005). Much of the relevant literature to date has focused on associations between child maltreatment and dysregulation of the HPA axis, with less attention to the ANS. The goal of the current review is to summarize and synthesize findings from the extant literature regarding ANS responsivity among children who have experienced maltreatment. A secondary goal is to explore the role of ANS responsivity in the known pathway between child maltreatment and psychopathology.

ANS Indicators and Functioning

The stress response system functions to coordinate an organism’s response to threats and encode information about the safety of the environment (Del Giudice, Ellis, & Shirtcliff, 2011). The ANS comprises two reciprocal branches, the sympathetic nervous system (SNS) and the parasympathetic nervous system (PNS; Figure 1). The SNS coordinates the “fight or flight” response, mobilizing an individual’s resources to respond to environmental demands (Porges, 2004). The PNS functions to reduce physiological arousal and to promote homeostasis, thereby supporting self-regulation, sustained attention, and social engagement (Del Giudice et al., 2011) and inhibiting sympathetic arousal (Porges, 2004). Under stress, PNS activity may be reduced, thus facilitating SNS activation to increase arousal. Following threat cessation, the PNS reasserts its influence over the SNS to reduce arousal and promote recovery.

Figure 1.

Description of the human autonomic nervous system stress response and measures of autonomic nervous system responsivity. N refers to the number of studies included in the systematic review utilizing each measure.

When the SNS is activated, heart rate becomes faster and more regular. Preejection period (PEP), the period between contraction of the ventricles of the heart and the ejection of blood into the aorta, is considered a “pure” measure of SNS activation, as it has been found to precisely mirror activation of heart contraction by the SNS (Schachinger, Weinbacher, Kiss, Ritz, & Langewitz, 2001). Indirect measures of SNS activation include heart rate, systolic blood pressure (maximum pressure during one heart beat), and diastolic blood pressure (minimum pressure in between two heart beats). SNS activation can also be measured through indices of skin conductance, which reflects arousal through SNS-controlled changes in the activity of the eccrine sweat gland (Dawson, Schell, & Filion, 2000). This arousal can be measured as tonic (skin conductance level [SCL]) or rapid, phasic changes (e.g., skin conductance response [SCR], galvanic skin response [GSR]). Thus, SNS activation may be indicated by measures of shortened PEP, accelerated heart rate, increased blood pressure, and/or increased SCL/SCR/GSR.

PNS activation both reduces heart rate and allows heart rate to become irregular, with heart rate increasing with inhalation and decreasing with exhalation. This respiratory system modulation of autonomic outflow to the heart and blood vessels—respiratory sinus arrhythmia (RSA)—is often used as an index of cardiac vagal activity. RSA under nonstress conditions (“baseline RSA”) is an index of vagal tone. “Vagal tone” indicates the degree to which the vagus nerve, which regulates homeostasis in the body, can cause the body to relax. “Vagal responsivity” reflects a change from baseline RSA to RSA under conditions of challenge, with vagal suppression or withdrawal reflecting decreased RSA from baseline to challenge. Thus, vagal withdrawal (i.e., decreased RSA) indicates reduced parasympathetic control over sympathetic activation to facilitate the mounting of a stress response to cope with challenge (Figure 1). Additionally, decreased heart rate and blood pressure may indirectly reflect parasympathetic activation.

Child Maltreatment and ANS Functioning

Systematic reviews and meta-analyses have demonstrated that individuals who have experienced child maltreatment have an increased risk throughout the life course for various forms of psychopathology, including post-traumatic stress disorder (PTSD), anxiety disorders, depressive disorders, disruptive behavior disorders, eating disorders, sleep disorders, substance misuse, and suicidality (e.g., Brown, 2003; Chen et al., 2010; Kendall-Tackett, Williams, & Finkelhor, 1993; Mulvihill, 2005; Paolucci, Genius, & Violato, 2001). However, not all maltreated children develop psychopathology (e.g., Cicchetti, 2010). While literature in this area is developing (e.g., see Afifi & MacMillan, 2011, for a review), more research is needed to determine why some children are resilient to these negative outcomes and others are not.

The extant literature suggests that differences in stress responsivity, including functioning of the ANS, may contribute to risk of psychopathology following maltreatment (McLaughlin, Sheridan, Alves, & Berry Mendes, 2014). First, children exposed to adversity are at increased likelihood for experiencing ANS dysregulation (e.g., Ellis, Essex, & Boyce, 2005; Miscovic, Schmidt, Georgiades, Boyle, & MacMillan, 2009). Second, ANS dysregulation is associated with psychopathology. For example, a meta-analysis conducted by Graziano and Derefinko (2013) demonstrated negative associations between vagal responsivity and internalizing and externalizing psychopathology in children. Less is known about the potential link between SNS activity and psychopathology, but patterns of low ANS responsivity have been found in children with externalizing problems (e.g., Crowell et al., 2006; Snoek, Van Goozen, Matthys, & Buitelaar, 2004; Van Goozen, Matthys, Cohen-Kettenis, Buitelaar, & Van Engeland, 2000). These findings are suggestive, but by no means conclusive, that disruptions to ANS functioning may mediate associations between child maltreatment and psychopathology. Additionally, El Sheikh (2005) found that marital conflict was associated with greater externalizing problems in girls who showed high sympathetic responsivity measured via skin conductance, suggesting that ANS responsivity may moderate associations between adversity exposure and psychopathology risk. Thus, there is limited evidence that ANS disruptions may serve as both a mediator and moderator of links between child maltreatment and psychopathology. Work is needed to explicate the exact nature of these associations.

Theoretical Constructs

At least two theoretical models may be relevant for considering the potential role of ANS functioning in the association between child maltreatment history and psychopathology risk. The differential susceptibility theory (DST; Belsky, Bakermans-Kranenburg, & van IJzendoorn, 2007) postulates that children vary in susceptibility to environmental influences (e.g., parental behaviors). Those most susceptible suffer the worst outcomes when exposed to poor/harmful parenting behaviors but the most optimal outcomes in response to positive parenting. These differences in susceptibility may be the result of genetic or other biologically based factors including differences in the responsivity of stress systems (e.g., Belsky et al., 2007). Thus, preexisting individual differences in ANS responsivity may influence children’s susceptibility to the negative effects of maltreatment. Following from this theory, ANS functioning moderates risk of psychopathology from maltreatment exposure.

The adaptive calibration model (ACM; Del Giudice et al., 2011) builds on the theory of biological sensitivity to context (e.g., Boyce & Ellis, 2005) to argue that individual differences in stress responsivity are the result of an individual’s adaptation to their environment. In this way, children adaptively respond to both unsupportive and protective family environments by modifying biologically and evolutionarily based stress response systems. The specific nature of a child’s calibration of the stress response depends in part on the quality of the child’s environment, particularly in early life when stress response systems are more plastic. Thus, in this model, early life experiences influence future susceptibility to environmental influences. Under highly stressful conditions, such as in the context of maltreatment, a very reactive stress response system (“hyperresponsivity”) may be most adaptive to allow the child to detect threat and act appropriately. Conversely, low responsivity of the stress response system (“hyporesponsivity”) may be adaptive to promote insensitivity to threat for children experiencing persistent, severe stress. Data suggest that the nature of maltreatment (i.e., whether maltreatment involves emotional, physical, and/or sexual abuse, and/or neglect) and the age at which the child experiences maltreatment influence whether stress systems become hypo- or hyperresponsive (Gunnar & Quevedo, 2007). Although potentially adaptive in the immediate context, both hypo- and hyperresponsive patterns may have long-term negative consequences (Gunnar & Quevedo, 2007; Parker, Buckmaster, Sundlass, Schatzberg, & Lyons, 2006). Hyperresponsivity may result in exposure to excessive levels of stress hormones, which may contribute to psychopathology (e.g., Staufenbiel, Penninx, Spijker, Elzinga, & van Rossum, 2013). Hyporesponsivity may also increase risk of poor health outcomes (Phillips, Ginty, & Hughes, 2013) including psychopathology (de Rooij, 2013). Following this theory, ANS reactivity may both be affected by the experience of maltreatment (i.e., mediate) and influence (i.e., moderate) the impact of later maltreatment on psychopathology risk.

The Current Review

The overall goal of this review was to systematically evaluate the evidence regarding ANS responsivity in maltreated children. All types of child maltreatment—physical, emotional, and sexual abuse and physical and emotional neglect—were considered, as the current literature is not developed well enough to indicate how different types of maltreatment may influence ANS outcomes (Gunnar & Quevedo, 2007). Furthermore, co-occurrence of different kinds of maltreatment is common (Cicchetti & Toth, 1995). Studies that included domestic violence (i.e., child witnessing or being a victim of domestic violence) occurring alongside other forms of abuse or neglect were also considered. Studies that assessed marital violence but not child maltreatment were excluded, as this was considered a distinct construct (i.e., not involving neglect or violence directed toward the child) from other forms of maltreatment (e.g., Saltzman, Holden, & Holahan, 2005). This review aimed to address the following question: Is a history of maltreatment associated with abnormalities of ANS responsivity in children? A secondary objective was to explore the role of ANS responsivity in the pathway to psychopathology among children who have experienced maltreatment. Studies were reviewed with attention to important developmental issues such as age at exposure and assessment.

Method

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were followed. Electronic searches of the following databases were conducted: Embase (Ovid; 1947–2019), Medline (Ovid; 1946–2019), ASSIA (1987–2019), PsycINFO (1887–2019), and CINAHL (1981–2019). The final search was conducted in March 2019. Key word searches utilizing the following terms were performed:

Child abuse OR abuse OR domestic violence OR child neglect OR child sexual abuse OR interpersonal violence OR maltreatment,

Cardiovascular response OR cardiovascular reactivity OR autonomic nervous system OR heart rate variability OR sinus arrhythmia OR pre-ejection period OR heart rate OR stress reactions OR vagal OR sympathetic OR parasympathetic,

Skin conductance OR galvanic skin response OR GSR OR skin conductance level OR SCL OR skin conductance response OR SCR.

Key word searches were also performed in relevant journals (Journal of Clinical Child and Adolescent Psychology, Journal of Biological Psychiatry, Journal of Biological Psychology, Child Abuse & Neglect, and Child Maltreatment), and several authors with expertise in the area were contacted to enquire about any missed studies or studies in press. Reference sections of articles were hand-searched to ensure that no relevant articles had been missed.

Articles were selected on the basis of meeting the following inclusion criteria:

Participants were children (0–19 years).

Childhood maltreatment was assessed via self- or parent/caregiver-report or outside agencies (e.g., child protection agencies).

Cardiovascular or skin conductance measures of ANS (PEP, RSA, heart rate, blood pressure, SCL, SCR, and/or GSR) were taken during a stressful task.

Where mediation and/or moderation was examined, associations between ANS responsivity and symptoms of psychopathology, including internalizing/externalizing symptoms and specific diagnoses, for example PTSD, were assessed.

Data were extracted from the selected studies. The quality of selected studies was assessed independently by two of the authors using the Crowe Critical Appraisal Tool (CCAT V1.4) . This tool includes scoring of items covering preliminaries, introduction, design, sampling, data collection, ethical matters, results, and discussion, resulting in a total score of 40 that indicates the assessed quality overall. According to the tool’s guidelines, a score of <20 is considered low quality, a score of 20–29 moderate quality, and a score of 30–40 high quality.

Results

The search produced a total of 1,388 articles (Figure 2). Articles were removed/excluded if they were duplicates (n = 204), conference proceedings or master’s theses (n = 10), or not about ANS responsivity or child maltreatment (n = 1,119). Abstracts for 55 articles were read, and those that clearly did not meet the inclusion criteria were excluded (n = 19), leaving 36 articles that were read in full. Twenty-one articles met the inclusion criteria. Additional hand searches and contacting experts yielded one additional article that met inclusion criteria, resulting in a total of 22 articles for inclusion. All included articles were assessed as being of high or moderate quality via the CCAT tool (i.e., received a score ≥20). There was good agreement between two independent ratings of the articles. Where there were differences in agreement, an agreed score was settled upon through conference.

Figure 2.

Inclusion and Exclusion of Articles.

Heterogeneity of Studies

Table 1 provides a summary of each study. Of the 22 studies, one tested moderation effects of ANS responsivity on the association between child maltreatment and psychopathology, and six tested whether ANS responsivity mediated the association between child maltreatment and psychopathology. One tested a moderated-mediation model for the association between ANS responsivity and psychopathology. Ten utilized generic cardiovascular responsivity measurements (heart rate or blood pressure), 10 utilized specific measures of SNS activity (PEP, SCR, or SCL), and 11 utilized specific measures of PNS activity (RSA; Figure 1).

Table 1.

Included Articles, Organized Alphabetically.

Authors/date	Study -design	Outcomes	Participants (n, Age, Recruitment)	Type of Maltreatment	Stimuli	Autonomic Measures	Type of ANS measures	Confounders included in analysis	Results	CCAT Score
Ben-Amitay et al. (2016)	Cross-sectional	Physiological responses to neutral and trauma-related stimuli among girls and women who experienced childhood sexual abuse	Sexually abused girls and women (n = 35) aged 7–51 years recruited from child abuse services and rape crisis centers and volunteer controls (n = 25)	Sexual abuse assessed via CTQ and interview with psychiatrist	Images that were positive, negative, neutral, or sexually allusive	HR SCL	Responsivity	Age	Heart rate and SCL were higher when viewing a sexually allusive image among abused girls and women compared to controls	28 Moderate
Busso et al. (2017)	Cross-sectional	Associations between threat (adjusting for deprivation) and deprivation (adjusting for threat) with autonomic and neuroendocrine stress responses Mediation of effects of childhood adversity on internalizing and externalizing symptoms by physiological responsivity	Adolescents (n = 169) mean age = 14.9 years recruited from schools, after-school clubs, medical clinics at Boston Children’s Hospital and the wider community	Emotional abuse, physical abuse, or sexual abuse assessed via CTQ Exposure to violence assessed via SAVE.	TSST	RSA PEP	Baseline Responsivity	AgeGender Poverty	Interpersonal violence, adjusting for poverty, was associated with blunted sympathetic responsivity Sympathetic responsivity did not mediate between violence exposure and psychopathology	34 High
Carrey et al. (1995)	Cross-sectional	Physiological responses of abused children to different stimuli compared to responses of a comparison group	Abused children (n = 18) age 7–13 years recruited from a Children’s Aid Society and a children’s hospital. Comparison group recruited from schools	Physical or sexual abuse that had been investigated legally and confirmed by medical/ psychiatric evaluations	Eight stimulus conditions (including relaxation and test conditions) presented on slides in front of child	Pulse height SCL	Baseline Responsivity	Gender	Abused children showed significantly lower pulse height at baseline, and smaller changes in pulse height from baseline to stressful stimuli (no signal and maths conditions), as well as significantly lower SCL compared to nonabused children	20 Moderate
Cipriano et al. (2011)	Cross-sectional	Relations between cardiac responsivity, family violence exposure and preschool children’s emotional adjustment	Mother–preschooler dyads (maltreated n = 41; nonmaltreated n = 33) children age 3–5 years recruited from Department of Public Welfare agencies and a database of birth announcements	Physical abuse or neglect or emotional maltreatment identified through CYS records	Mother–child joint duplo task. Individual transparent box task. Individual shape and day/night tasks.	RSA	Baseline Responsivity	Age Gender	Among children living in violent contexts (index for child maltreatment) vagal withdrawal during challenge was unrelated to their emotional adjustment	30 High
Ford et al. (2010)	Cross-sectional	Association of stress-related changes in ANS responsivity among children with a history of physical/sexual abuse	Pediatric psychiatric inpatients (abused n = 224; nonabused n = 38) mean age = 13 years recruited from a residential treatment center for high risk and emotionally disturbed children/adolescents	Physical or sexual abuse documented by CPS	Clinical venepuncture	HR DBP SBP	Baseline Responsivity	Age Gender BMI Verbal IQ Performance IQ	Participants who exhibited slower HR compared to baseline following stressor were more likely to have a history of physical abuse than those who showed no change or faster HR	32 High
Gunnar et al. (2009)	Cross-sectional	Impacts of moderate and severe early life stress on the HPA axis	Early adopted/foster care children (n = 44) and later adopted/postinstitutionalised children (n = 42) recruited from the Minnesota International Adoption Project Nonadopted children (n = 38) recruited from the Institute of Child Development Participant Pool Age 10–12 years	Likely neglect as a result of institutional care; Child Life Events Scale completed by carers	TSST	PEP RSA	Baseline Responsivity	Age BMI Family income Parent/carer’s education	Later adopted/postinstitutionalized children (but not early adopted/foster care children) had lower overall PEP compared to nonadopted children	30 High
Heleniak et al. (2016)	Cross-sectional	Impact of childhood trauma on cardiovascular stress reactivity	Adolescents (n = 488) participating in the TRacking Adolescents’ Individual Lives Survey mean age = 16.17 years	Sexual abuse or other traumatic experience including natural disasters, life-threatening accidents, witnessing severe injury/death of another person, or being held captive/abducted assess via self-report questionnaire Family violence/physical abuse assessed via the CTS	GSST	PEP	Baseline Responsivity Recovery	Gender Age Maternal education	Exposure to trauma was associated with internalizing and externalizing symptoms Adolescents who had been exposed to trauma exhibited blunted cardiac reactivity during the stress task Blunted cardiac reactivity was positively associated with externalizing symptoms and mediated the link between trauma and externalizing psychopathology	33 High
Hill et al.(1989)	Cross-sectional	Comparison of abused and nonabused children in their perception of everyday events and their physiological response to everyday events	Hospitalized abused (n = 29) and nonabused (n = 20) children age 7–15 years recruited from a psychiatric hospital Nonhospitalised children (n = 50) age 7–12 years recruited from parents at the University of New Orleans	Physical abuse that had been documented and resulted in children being removed from parents	Video Apperception Test	HR	Baseline Responsivity	Age Gender	Abused children showed slower HR compared to baseline in response to scenes of conflict/fearful situations	25 Moderate
Jennes et al. (2018)	Cross-sectional	Fear extinction learning as a mechanism underlying protective effect of vagal tone on PTSD symptoms among abused children	Abused (n = 38) and nonabused (n = 56) children age 6–18 years recruited from schools, after-school clubs, medical clinics and the general community	Physical abuse, sexual abuse, and witnessing domestic violence assessed via CECA and CTQ	Fear conditioning task validated for children	RSA SCR	Baseline Responsivity	Age Gender Socioeconomic status	High RSA associated with lower PTSD symptoms among abused children. This association was mediated by enhanced fear extinction learning (lower SCR during early extinction learning)	35 High
Koopman et al. (2004)	Cross-sectional	Relationship of dissociative symptoms, abuse and neglect, and gender to heart rate in stressful and nonstressful interviews	Children (n = 41) age 11–16 years recruited from a juvenile probation department	Physical abuse, sexual abuse, emotional abuse, physical neglect, and emotional neglect, assessed through the CTI	Stressful event interview or free association interview	HR	Responsivity	Age Gender	Mean HR was faster among those who reported more abuse or neglect Faster mean HR was associated with participation in the free association interview	30 High
Leitzke et al. (2015)	Cross-sectional	ANS responsivity to a laboratory-based performance/peer rejection stressor in maltreated and nonmaltreated children	Children (maltreated n = 34; nonmaltreated n = 77) age 9–14 years recruited via advertisements in the community.	Type not identified, maltreatment verified through score of >20 on the Parent–Child CTS or reports of abuse recorded with Dane County Department of Human Services	Surprise speech task followed by negative peer feedback via computer	HR SBP DBP	Baseline Responsivity	Age	SBP was lower for maltreated children poststressor than for nonmaltreated children DBP was lower for maltreated children at baseline and poststressor than for nonmaltreated children	35 High
Lunkenheimer et al. (2019)	Cross-sectional	Covarying mother and child individual and dyadic RSA with interactive repair in maltreating and nonmaltreating dyads	Maltreating (n = 59) and nonmaltreating (n = 42) mother–children dyads. Children age 3–5 years. Recruited from a study on parent–child interaction and child maltreatment	Physical abuse or neglect assessed via CPS records.	Two dyadic puzzle tasks, including a puzzle of a train, and constructing a replica of a three-dimensional figure out of Duplo Lego	RSA	Baseline Responsivity	Total utterances made Maternal age Maternal income	Lower levels of mother repair associated with decreases in children’s RSA in nonmaltreating dyads but increases in children’s RSA in maltreating dyads	33 High
Macmillan et al. (2009)	Longitudinal	Differences in heart rate and cortisol resting and responsivity levels in response to a psychosocial stressor between maltreated and nonmaltreated adolescents	Maltreated female children (n = 67) age 12–16 years recruited from child protection agencies Age-matched control group (n = 25) recruited from an existing research database	Physical abuse, sexual abuse, emotional abuse, emotional neglect, witnessing domestic violence, identified through case review of files from Child Protection Agencies	TSST	HR	Baseline Responsivity Recovery	Time of day	Both maltreated and nonmaltreated girls showed faster HR compared to baseline following the TSST and a gradual decrease over time	33 High
McLaughlin et al. (2014a)	Cross-sectional	Vagal tone and vagal responsivity following psychosocial stressors and their influence on psychopathology among adolescents exposed to childhood adversity	Adolescents (maltreated n = 60; nonmaltreated n = 97) age 13–17 years recruited from schools, after school programs, medical clinics and the general community	Physical abuse, emotional abuse, sexual abuse, assessed through CTQ and SAVE	TSST	RSA	Baseline Responsivity	Age Gender Respiration rate	Significant positive association between child abuse and internalizing problems for adolescents with low baseline vagal tone Adolescents who had experienced child abuse and had low RSA withdrawal (vagal withdrawal) during the TSST were more likely to have internalizing problems	32 High
McLaughlin et al. (2014b)	Cross-sectional	Stress responsivity following child maltreatment	Adolescents (n = 60; nonmaltreated n = 97) age 13–17 years recruited from schools, after school programs, medical clinics and the general community	Physical abuse, sexual abuse, emotional abuse, assessed through the CTQ and CECA	TSST	BP PEP	Baseline Responsivity	Age Gender Single-parent household Poverty	During the maths component of the TSST, maltreated adolescents exhibited significantly less PEP responsivity (less decreases compared to baseline) than nonmaltreated adolescents	35 High
McLaughlin et al. (2015)	Longitudinal	Development of the ANS and HPA axis in children exposed to early-life deprivation associated with institutional care	Part of the Bucharest Early Intervention Project. Children from Romanian institutions randomized to a high-quality foster care intervention (n = 48) and children remaining in Romanian institutions (n = 43) Typically developing Romanian children (n = 47). Mean age = 12 years	Likely neglect as a result of institutional care	TSST Frustration task	HR SBP DBP PEP RSA	Baseline Responsivity	Pubertal development	Children remaining in institutional care showed blunted HR, DBP and PEP during the speech component of the TSST, and blunted DBP and PEP during the maths component	37 High
McLaughlin et al. (2016)	Cross-sectional	Relation of trauma and neural structure to fear conditioning in children	Maltreated children aged 6–18 years (n = 38) and controls (n = 56) recruited from schools, after school and prevention programs and medical clinics	Physical abuse, sexual abuse, or domestic violence assessed via the CECA and CTQ	Fear conditioning paradigm using bells as the conditioned stimulus and an aversive alarm noise as the unconditioned stimulus	SCL	Responsivity	Age Gender IQ Poverty Nonmaltreatment trauma exposure Symptoms of anxiety and depression PTSD symptoms PTSD diagnosis Externalizing problems	Maltreated children exhibited blunted SCL to conditioned stimulus compared to controls and failed to show differential SCL to threat cues compared to safety cues during conditioning. This altered pattern of conditioning mediated the relationship between maltreatment and externalising psychopathology	36 High
Oosterman et al. (2010)	Cross-sectional	ANS activity of children during separation and reunion with foster carers	Foster children (n = 60) age 2–7 years and their primary foster parent	Neglect, physical abuse, sexual abuse, witnessing domestic violence, identified through questionnaires completed by child welfare case workers	Strange Situation (modified for older children)	PEP RSA	Responsivity	Age Time in placement Previous placements	Children with higher maltreatment scores showed less PEP responsivity (less decreases compared to baseline) during first separation. Sexually abused children showed decreased RSA (vagal withdrawal) on separation and increased RSA on reunion compared to children without a background of sexual abuse	28 Moderate
Pollak et al. (2005)	Cross-sectional	Attentional regulation in response to interpersonal anger among physically abused children	Abused (n = 11) and nonabused (n = 22) children age 4–5 years recruited through Department of Human Services and flyers posted locally	Physical abuse, cases substantiated by Dane County Department of Human Services	Background conversation including baseline period, active anger period, unresolved anger period and resolution period	HR SCL	Baseline Responsivity Recovery	Age Gender	Abused children showed slower HR compared to baseline in response to the anger period and maintained this state and showed a steady decline in SCL across the anger period. Nonabused children were initially aroused (slower HR compared to baseline and increase in SCL) in response to anger but showed recovery to baseline when the conflict was resolved	36 High
Shenk et al. (2010)	Longitudinal (assessed at six time points over 13 years)	Stress response in maltreated children and levels of psychopathology over time	Abused females (n = 52) age 6–16 years recruited from CPS agencies. Comparison group recruited from community advertisements (n = 77). Assessments providing this data took place at age 18 and 24 years	Sexual abuse, substantiated by CPS agencies in Washington, DC.	Timed mental rotation task; incorrect responses followed by an aversive noise	RSA	Baseline Responsivity	Age Minority status Previous levels of internalising and externalising behaviours	Sexual abuse in childhood significantly predicted an asymmetrical physiological response (blunted cortisol response and vagal withdrawal) to stressor at late adolescence (18 years)	37 High
Shenk et al. (2012)	Cross-sectional	Relationship between RSA and cortisol responsivity, experiential avoidance, PTSD and child maltreatment	Adolescent females (maltreated n = 51; nomaltreated n = 59) age 14–19 years recruited from CPS agencies and teen health centers	Physical neglect, physical abuse or sexual abuse determined by CPS agency investigation.	Performance (timed affect recognition task) and interpersonal (video clips of parent-adolescent conflict) tasks	RSA	Baseline Responsivity	Age Cigarette use Steroid use	Vagal responsivity alone did not significantly mediate the relationship between child maltreatment and PTSD	36 High
Shenk et al. (2014)	Longitudinal (1 year follow up)	See above	See above	See above	See above	RSA	Baseline Responsivity	See above	See above	33 High

Note. ANS = autonomic nervous system; BP = blood pressure; BMI = body mass index; CCAT = Crowe Critical Appraisal Tool; CECA = childhood experiences of care and abuse; CPS = Child Protection Service; CTI = comprehensive trauma interview; CTQ = Childhood Trauma Questionnaire; CTS = Conflict Tactics Scale; CYS = Children and Youth Services; DBP = diastolic blood pressure; GSST = Groningen social stress task; HR = heart rate; PEP = preejection period; PTSD = post-traumatic stress disorder; RSA = respiratory sinus arrhythmia; SAVE = screen for adolescent violence exposure; SBP = systolic blood pressure; SCL = skin conductance level; SCR = skin conductance response; TSST = Trier social stress test; IQ = intelligence quotient; BMI = body mass index.

The age of participants in the included studies ranged from 2 years to 19 years. Seventeen studies investigated physical abuse, five studies examined physical neglect, 17 studies examined sexual abuse, six studies examined emotional abuse, and three studies examined emotional abuse or neglect. Four studies included witnessing domestic violence, and one study did not specify the type of maltreatment experienced by participants. Two studies drew samples from institutionalized children, where neglect was presumed to have occurred. Nineteen studies included a nonmaltreated comparison group, and three studies compared ANS responsivity within maltreated samples.

Reflective of the wide age range of participants in the included studies, the type of stimuli used to assess participants’ stress responses also varied, including physical, emotional, and cognitive challenges. One study used a clinical venipuncture; two a fear conditioning paradigm; one images that were positive, negative, or sexually allusive; one an interview asking participants to describe a stressful event or a free association task; one used the video apperception test (during which participants watched video clips of scenes depicting everyday situations or conflict between a child and an adult and answered questions about them); one used a modified strange situation procedure; one presented children with relaxation and test conditions (e.g., math calculations) via slide projections; two used an affect recognition task and video clips depicting conflict; one used a background conversation including periods of active and unresolved anger; one used mother–child joint tasks and individual tasks representing cognitive and emotional challenges; one used mother–child joint puzzle tasks; one used a timed mental rotation task (during which participants determined whether two rotated stimuli were the same objects or one was an inverted image of the other); and eight used a version of the trier social stress test (TSST; a structured set of tasks that include making a speech).

As the goal of this review was to take first steps to evaluate the state of the literature, a meta-analysis was outside the scope of this review. A narrative synthesis of findings is therefore presented in the following section.

Studies Examining Cardiovascular Responsivity and Maltreatment

The majority of the studies examining cardiovascular responsivity demonstrated that maltreated children exhibited a blunted response in the form of diminished SNS activity compared to nonmaltreated children. Carrey, Butter, Persinger, and Bialik (1995) compared physiological responses to relaxation and test conditions among children aged 7–13 years who had been physically or sexually abused and a nonabused control group. Compared to nonabused children, abused children exhibited significantly lower pulse height at baseline and smaller changes in pulse height from baseline to test conditions. Hill, Blechfield, Brunstetter, Herbert, and Steckler (1989) measured heart rate in 7- to 15-year-old physically abused and nonabused children during the video apperception test and found that physically abused children showed slower heart rate compared to baseline in response to video scenes depicting conflict or fearful situations. Leitzke, Hilt, and Pollak (2015) measured ANS activity while children aged 9–14 years participated in a surprise speech task. Compared to nonmaltreated children, maltreated children exhibited lower systolic blood pressure after the stressor and lower diastolic blood pressure at both baseline and poststressor. Pollak, Vardi, PutzerBechner, and Curtin (2005) gave physically abused and nonabused children aged 4–5 years a task to complete while a conflict conversation that included periods of active and unresolved anger played in the background. During the conflict and resolution, physically abused children showed slower heart rate compared to baseline, which the authors interpreted as indicating attentional arousal in response to the active anger, unresolved anger, and resolution periods. These findings suggest maintenance of a state of anticipatory monitoring throughout the conversation. This contrasts to the response of nonabused children, who showed initial slowed heart rate compared to baseline in response to anger but then recovered to baseline when the conflict was resolved. Finally, Ford, Fraleigh, Albert, and Conor (2010) exposed pediatric psychiatric inpatients aged 13 years to a clinical venepuncture. Those who exhibited slower heart rate relative to baseline following the stressor were more likely to have a history of physical abuse, as opposed to sexual abuse or no abuse, than those who showed no change or an increase in heart rate.

In contrast to these findings, two studies demonstrated either heightened cardiovascular responsivity among maltreated children or no differences between maltreated and nonmaltreated children. Koopman and colleagues (2004) exposed maltreated children aged 11–16 years from a juvenile probation center to stressful and nonstressful interviews and found that heart rate during these interviews varied with magnitude of maltreatment exposure: Mean heart rate during both interviews was faster among those who had higher scores on the childhood trauma interview, which assessed physical, sexual, and emotional abuse and physical and emotional neglect. In addition, in a longitudinal study of female children aged 12–16 years, MacMillan and colleagues (2009) found no differences in baseline or responsivity levels of heart rate following a social stress task between those who had experienced maltreatment (physical, sexual, or emotional abuse; emotional neglect; or witnessing domestic violence) and those who had not. Both groups demonstrated faster heart rate compared to baseline following the task and then a gradual decrease over time.

Studies Utilizing Specific Measures of SNS Activity in Association With Maltreatment

Six studies examined PEP in association with maltreatment. Five of these studies demonstrated that maltreated children exhibited blunted sympathetic activation during stressors compared to nonmaltreated children. Busso, McLaughlin, and Sheridan (2017) administered a social stress task to adolescents and found that those who were exposed to interpersonal violence (emotional abuse, physical abuse, and/or sexual abuse) exhibited blunted SNS responsivity during the speech and math components of the test compared to those who were not exposed to interpersonal violence. Gunnar, Frenn, Wewerka, and Van Ryzin (2009) demonstrated that children aged 10–12 years who had been cared for predominantly in orphanages had lower overall PEP scores compared to children who had been adopted early in life and children who lived with their birth families and were therefore presumed to not have experienced maltreatment. No change in PEP relative to baseline occurred during a social stress task for any group of children. Heleniak, McLaughlin, Ormel, and Riese (2016) measured PEP in adolescents while they completed a social stress task. Greater exposure to trauma including sexual abuse, physical abuse, or another traumatic event (e.g., natural disasters, being held captive) was associated with blunted decreases in PEP responsivity during the speech component of the task. McLaughlin, Sheridan, Alves, and Berry Mendes (2014) measured cardiac output during a stress task in 13- to 17-year-old adolescents who had experienced physical, sexual, and/or emotional abuse and in nonmaltreated controls. Exposure to maltreatment was associated with less PEP responsivity (i.e., smaller decrease compared to baseline) during the math component of the task. McLaughlin and colleagues (2015) found that 12-year-old children in lifetime institutional care in Romania showed blunted SNS activation, including heart rate, diastolic blood pressure, and PEP, in response to a social stress task compared to children in foster care. The sixth study examined severity of maltreatment in relation to SNS activation. Oosterman, De Schipper, Fisher, Dozier, and Schuengel (2010) exposed children aged 2–7 years who had experienced physical or sexual abuse or witnessed domestic violence to an adaptation of the strange situation procedure with their foster carers. Children with higher risk scores, indicating more severe maltreatment, demonstrated less PEP responsivity during the first separation from their foster carer than did those with lower risk scores.

Studies Examining Skin Conductance and Maltreatment

Five studies examined skin conductance in association with maltreatment. Among these studies, three measured SCL, two measured SCR, and one measured GSR. Reported associations between skin conductance and maltreatment were mixed, with three studies reporting blunted response among maltreated children compared to controls, and one study reporting no differences between maltreated and control children. McLaughlin and colleagues (2016) exposed maltreated children and nonmaltreated controls aged 6–18 years to a fear conditioning paradigm. During the conditioning phase, maltreated children showed a blunted SCR to the conditioned stimulus. Carrey and colleagues (1995) found that maltreated children showed lower GSR than a community sample of children across relaxation and challenge stimulus conditions. Additionally, maltreated children demonstrated smaller changes in GSR from baseline than controls. Pollak and colleagues (2005) reported that when 4- to 5-year-old children were exposed to a periphery argument while completing a task, abused children showed blunted SCL, while nonabused controls demonstrated increased SCL. Jenness, Bryant Miller, Rosen, and McLaughlin (2018) exposed abused and nonabused children to a fear conditioning paradigm, finding that abused children who demonstrated high levels of resting RSA showed lower SCR during extinction learning, while the nonabused group of children showed lower SCR during extinction learning among children with low resting RSA. One study did not find any associations between maltreatment and skin conductance. Ben-Amitay, Kimchi, Wolmer, and Toren (2016) found no differences in SCR between maltreated and nonmaltreated children viewing video stimuli (e.g., negative, sexually allusive).

Studies Utilizing Specific Measures of PNS Activity in Association With Maltreatment

Three studies examined RSA in association with maltreatment, with mixed findings. Oosterman, De Schipper, Fisher, Dozier, and Schuengel (2010) exposed children aged 2–7 years who had experienced physical or sexual abuse or witnessed domestic violence to an adaptation of the strange situation procedure with their foster carers. Children who had experienced sexual abuse showed decreased vagal withdrawal (i.e., smaller decreases in RSA from baseline to challenge) on separation and increased vagal withdrawal on reunion compared to children without a history of sexual abuse. Shenk, Noll, Putnam, and Trickett (2010) examined physiological responses to a timed mental rotation task in sexually abused and nonabused 18-year-old females. Those who had experienced sexual abuse exhibited an asymmetric physiological response to the task, characterized by vagal withdrawal and a blunted cortisol response. Lunkenheimer, Busuito, Brown, Panlilio, and Skowron (2019) examined covariation of mother–child individual and joint RSA with interactive repair during dyadic puzzle tasks in maltreating and nonmaltreating dyads. Low levels of mother repair were associated with increases in child RSA in maltreating dyads, compared to decreases in child RSA in nonmaltreating dyads.

ANS Responsivity as a Potential Mediator of Psychopathology Following Maltreatment

Six studies assessed ANS responsivity as a potential mediator of the association between maltreatment and psychopathology, including externalizing/internalizing problems, emotional adjustment, and PTSD. Findings regarding the role of sympathetic responsivity in mediating child maltreatment and psychopathology were varied. In a cross-sectional sample of adolescents who completed a social stress task, Busso, McLaughlin, and Sheridan (2017) did not find evidence for a mediating role of sympathetic responsivity between child violence exposure, including child maltreatment, and internalizing and externalizing symptoms. In contrast, in a cross-sectional study, Heleniak, McLaughlin, Ormel, and Riese (2016) found that blunted sympathetic responsivity among adolescents who completed a social stress task mediated the link between exposure to trauma in childhood (including sexual or physical abuse) and externalizing symptoms. Additionally, in a cross-sectional study, McLaughlin and colleagues (2016) found that, when exposed to a fear-conditioning paradigm, maltreated children, relative to nonmaltreated controls, showed blunted SCL responsivity to threat cues during fear conditioning and a lack of differential SCL responsivity to threat and safety cues during early conditioning. This altered fear conditioning pattern mediated the relationship between maltreatment and externalizing psychopathology.

A seventh cross-sectional study conducted by Jenness and colleagues (2018) tested a moderated-mediation model. The authors exposed abused and nonabused children to a fear-conditioning paradigm and found that among abused children, low SCR during early extinction learning mediated the association between high vagal tone and low levels of PTSD.

No studies demonstrated that vagal withdrawal during stress mediated the relationship between child maltreatment and psychopathology. Cipriano, Skowron, and Gatze-Kopp (2011) gave a cross-sectional sample of preschool children individual or joint challenge tasks to complete with their mothers. Among children living in violent contexts, vagal withdrawal during challenging tasks was unrelated to emotional adjustment. Shenk, Putnam, and Noll (2012) and Shenk, Putnam, Rausch, Peugh, and Noll (2014) administered a stressor paradigm to adolescent females who had experienced physical or sexual abuse or physical neglect to investigate mediators of the relationship between child maltreatment and PTSD 1 year later. Vagal responsivity alone did not significantly mediate this relationship. Rather, experiential avoidance—defined as an unwillingness to experience aversive private events, such as unwanted memories, and attempts to control or suppress such events (Hayes, Wilson, Gifford, Follette, & Strosahl, 1996)—was the only significant mediator of the relationship between child maltreatment and the development of PTSD symptoms.

ANS Responsivity as a Potential Moderator of the Effects of Maltreatment on Psychopathology

One study found evidence that ANS responsivity moderates the association between child maltreatment and internalizing symptoms. In a sample of adolescents who had or had not experienced physical, sexual, or emotional abuse, McLaughlin, Alves, and Sheridan (2014) assessed vagal tone at rest and vagal withdrawal in response to a social stress task. A positive association was found between child abuse exposure and internalizing problems for adolescents with low vagal tone and low vagal withdrawal during the stressor.

Discussion

The purposes of this review were to evaluate the evidence for disruptions in ANS functioning in maltreated children and to explore the role of ANS responsivity in the pathway from maltreatment to psychopathology. The majority of studies reported a similar pattern of ANS responsivity in maltreated children in the form of blunted cardiovascular/SNS responsivity during a stress-related/challenging task. Mixed findings were demonstrated for PNS activity, with one study finding that maltreated children showed less vagal withdrawal during separation from their caregivers but increased vagal withdrawal on reunion (Oosterman, De Schipper, Fisher, Dozier, & Schuengel, 2010) and another finding a more typical vagal withdrawal response to a challenging task in maltreated children (Shenk, Noll, Putnam, & Trickett, 2010).

Evidence for ANS functioning as a mediator or moderator of child maltreatment effects on psychopathology risk were mixed and limited. One study found ANS responsivity to be a potential moderator of the effects of child maltreatment on the risk of internalizing problems (McLaughlin, Alves, & Sheridan, 2014a). Studies investigating ANS responsivity as a mediator of child maltreatment on psychopathology reported inconsistent findings. Two studies (Heleniak, McLaughlin, Ormel, & Riese, 2016; McLaughlin et al., 2016) found evidence that blunted SNS responsivity may mediate the association between childhood trauma, including maltreatment, and psychopathology, while four studies found no mediating role for ANS responsivity (Busso, McLaughlin, & Sheridan, 2017; Cipriano, Skowron, & Gatze-Kopp, 2011; Shenk, Putnam, & Noll, 2012; Shenk, Putnam, Rausch, Peugh, & Noll, 2014). Interestingly, the two studies finding support for mediation effects specifically linked blunted SNS responsivity to externalizing symptoms, suggesting specificity in the associations between direction of ANS dysfunction and type of psychopathological symptoms. Notably, Heleniak, McLaughlin, Ormel, and Riese (2016) grouped a range of traumatic experiences (e.g., being involved in a natural disaster, being held captive) alongside exposure to child maltreatment. Thus, caution must be used when interpreting these results regarding the role of ANS responsivity in mediating child maltreatment effects. One study (Jenness et al., 2018) found evidence for a moderated-mediation model in which extinction learning (indexed via SCR) mediated the association between abuse and PTSD symptoms only among children with high resting RSA, suggesting that extinction learning may be a mechanism underlying the protective effects of high vagal tone in this population. Importantly, six of the seven studies testing mediation were cross-sectional (Table 1), a serious limitation. The findings from these studies should therefore be approached with caution and demonstrate a need for research that utilizes longitudinal data to properly test mediation effects of ANS responsivity in maltreatment–psychopathology associations.

Although some support was found for both the DST and ACM theories, the support is limited by inconsistencies in the findings and the study designs. DST hypothesizes that individual differences in ANS functioning may buffer or exaggerate the impact of child maltreatment effects on psychological functioning, such that the same maltreatment exposure may lead to varying levels of psychopathology depending on the individual’s preexposure ANS responsivity tendencies. ACM posits that children’s ANS functioning is influenced by their early experiences of child maltreatment and their resultant ANS functioning may influence the impact of later maltreatment exposures on psychopathology risk; thus, ACM may explain both mediation and moderation findings. Appropriate application of these models requires the study of very young children, beginning prior to maltreatment exposure, and longitudinal tracking to determine whether maltreatment exposure is independent of or increases risk of ANS dysregulation and whether ANS dysregulation modifies the impact of child maltreatment on psychopathology risk. As some of the study samples included in this review were adolescents, application of these models must be tempered, and further research with younger samples is required to understand how the ACM can explicate the nature of the relationship between child maltreatment, ANS responsivity, and psychopathology. Such research would need to undertake complex statistical approaches, given that the ACM predicts both hyper- and hypostress responsivities in contexts of stress, and these contrasting physiological responses might “cancel out” in traditional analyses that assume linear relationships. Moreover, the ACM assumes both mediation and moderation effects, with the nature of effects varying by time and nature of maltreatment exposure (e.g., early vs. later childhood, initial vs. repeated maltreatment exposure). Finally, other theoretical models not discussed here may contribute to our understanding of the associations among child maltreatment, ANS functioning, and psychopathology and may deserve consideration.

Findings from this review can be compared to others assessing constructs closely linked to maltreatment. For example, El Sheikh and Erath (2011) reviewed the literature on child ANS functioning in the context of family conflict and found that higher vagal tone and increased vagal withdrawal are protective factors among exposed children. For children living in such contexts, greater reactivity of the PNS may be adaptive given that the PNS plays a role in supporting emotion regulation and social engagement (Porges, 2007). However, the studies reviewed here found greater evidence for blunting in reactivity of both SNS and PNS among maltreated children, with maltreatment associated with blunted cardiovascular/sympathetic responsivity and lower vagal withdrawal in response to challenge. Indeed, Cipriano, Skowron, and Gatze-Kopp (2011) suggested that extreme family violence, such as maltreatment, may overpower the ability of children’s ANS to adapt flexibly to their environment. Given the state of the extant literature, the current review suggests that the role of ANS responsivity in the path from child maltreatment to psychopathology cannot yet be determined and requires further exploration.

Also relevant to consider is the different psychological/behavioral functions that may be served by different indices of ANS reactivity. For example, measures of SCL have been hypothesized to be particularly relevant to punishment sensitivity/valuation (e.g., Matthys, van Goozen, Snoek, & van Engeland, 2004), whereas measures of PEP may be particularly sensitive to reward valuation (Richter & Gendolla, 2009). Consideration of these differences in function may be particularly important when studying the role of ANS functioning in the association between child maltreatment and psychopathology risk. Future studies should thus consider the constructs hypothesized to be most relevant for the particular associations of interest to be examined (e.g., punishment avoidance in relation to externalizing symptom risk) and choose ANS indices that most closely tap into those constructs.

The inconsistencies in findings across studies are likely attributable in part to differences in study samples, methods, and statistical analyses. Consideration of different theoretical models, such as DST and ACM, may be helpful in informing future study designs. For example, DST may be more applicable to understanding how genetic differences in ANS responsivity may influence psychopathology risk in the context of maltreatment exposure, whereas ACM may be better suited for explaining the complex roles of chronic stressful experiences in both shaping ANS responsivity profiles and affecting psychopathology risk across childhood. Relevant here may be the discordant findings on the effects of child maltreatment on HPA axis functioning noted in the literature. Nemeroff (2016) suggested participant characteristics that may explain inconsistencies, including the type of maltreatment experienced, the presence/absence of psychosocial support, family history of psychiatric disorders, and genetic/epigenetic factors. The age at which the child was first exposed to maltreatment and the severity and chronicity of maltreatment exposure are also critical factors to consider. All these factors likely influence associations between child maltreatment history and ANS responsivity but have yet to be considered sufficiently in study designs. For example, not all of the reviewed studies clarified the nature of maltreatment experienced by participants or verified the occurrence of child maltreatment with child protection agencies (Table 1). The ages of participants at assessment varied across studies, from 2 to 19 years, without consideration of age at exposure. Studies are needed that examine the impact of age at exposure throughout childhood and adolescence, given that there may be multiple sensitive periods for exposure effects on ANS functioning. Notably, research has identified age and gender differences in children’s cardiac physiology (Fabes, Eisenberg, Karbon, Troyer, & Switzer, 1994; Quas, Hong, Alkon, & Boyce, 2000). Thus, normative developmental changes in ANS structure and functioning may influence the nature and magnitude of child maltreatment impact. In the reviewed studies, 19 controlled for age in analyses (Table 1), and of the 17 studies that included both male and female participants, 12 controlled for gender (Table 1). Future study designs should consider potential moderating effects of age and gender in their models to determine whether the impact of maltreatment exposure on ANS functioning and psychopathology differs by child age and/or by gender.

A significant limitation of the study findings is the lack of gold standards for assessing ANS responsivity, including the ideal design for stress tasks. The utilization of varied measures and methods for assessing ANS responsivity hinders efforts to summarize findings across studies and to define “abnormalities” in ANS functioning. Indeed, the varying nature of the stress tasks undertaken by participants across studies may have contributed to discrepancies in the results. As child maltreatment is often interpersonal in nature, interpersonal stress tasks like the TSST may be more likely to evidence differences in maltreated children’s stress responsivity. While the majority of studies (n = 16) made use of tasks that were at least in part interpersonal in nature, six did not (Table 1). In addition, Cacioppo, Uchino, and Bernston (1994) have highlighted the difficulty in teasing apart measurement of SNS from PNS activity, particularly when relying on heart rate and blood pressure. These measures are often used as indices of sympathetic activity despite also being influenced by the PNS (Figure 1). As such, it is important to consider that the blunted SNS response demonstrated in some of the described studies may in actuality reflect higher PNS activity, or both. In addition, baseline levels of autonomic arousal may be elevated in maltreated children, complicating analysis of changes in ANS activity in response to stress. Although the majority of studies assessed both baseline ANS functioning and stress responsivity, four studies did not (Table 1). Thus, it is difficult to establish a benchmark level of ANS responsivity that indicates that a task has been experienced as stressful for participants. A lack of change in ANS indices from baseline to stressor may indicate that the participant did not experience the task as stressful. If a task is stressful, a blunted ANS response may reflect an inability to mobilize resources to cope with threat. Exaggerated ANS responsivity could reflect hypersensitivity to threat in the environment. Resolution of methodological differences may help limit discrepancies in findings across studies. Additionally, standardizing methodologies and providing access to raw data within data repositories would facilitate the conduction of meta-analyses to further advance the field.

Recommendations for Future Work

As noted above, there is no consensus regarding the optimal task(s) to use for testing ANS responsivity, particularly with maltreated samples. The most common stress task used in the studies reviewed was the TSST, but this task cannot be used with young children due to the demands of the task. More research is needed to determine the best stress protocol(s) to use to measure ANS responsivity in maltreated children across different developmental stages. Protocols should be relevant for the populations of study and may vary, depending on the type of maltreatment as well as developmental stage. Importantly, the protocols must be ethical in light of the children’s prior maltreatment experience.

The extant literature is very limited regarding the role of ANS responsivity in the association between child maltreatment and psychopathology. More research is required to assess how ANS responsivity may mediate and/or moderate associations between child maltreatment and psychopathology. Study designs should be well informed by the developmental literature, taking into account factors the literature suggests may influence the magnitude and nature of these associations (e.g., child gender, age at exposure, type of psychopathology). Research that better characterizes samples for potential confounders is particularly needed. Careful measurement of factors that often covary with maltreatment and that may also contribute to ANS disruptions (e.g., socioeconomic status; other stress exposures such as family conflict, El Sheikh & Erath, 2011) should be considered in future research. Also, the cross-sectional design of most studies hampers efforts to determine mechanisms of effect of child maltreatment on ANS responsivity across the life span and prevents the drawing of conclusions regarding directions of effects. Rigorous longitudinal research is needed to address these issues. By evaluating the current state of the literature, this review provided a first step toward understanding the impact of child maltreatment on ANS functioning and the role of ANS dysregulation in the association between child maltreatment and psychopathology. A suggested next step would be to determine whether the disparate methodologies across studies allow for a meta-analysis to determine effect sizes and to spur additional research to address existing gaps.

Conclusion

This review suggests that ANS responsivity may be disrupted among maltreated children. Further, disruptions to ANS functioning may influence risk of psychopathology among maltreated children. As such, ANS responsivity may have important implications for intervention and treatment. As this review only examined cardiovascular and skin conductance measures of ANS responsivity, there are other issues that must be considered to develop an accurate and complete picture of stress responsivity and psychopathology in maltreated children. The current literature is far from conclusive, and much more work is needed to inform our understanding of these issues.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Genevieve Young-Southward

References

Afifi

T. O.

MacMillan

H. L.

(2011). Resilience following child maltreatment: A review of protective factors. The Canadian Journal of Psychiatry, 56, 266–272.

Alink

Cicchetti

Kim

(2012). Longitudinal associations among child maltreatment, social functioning and cortisol regulation. Developmental Psychology, 48, 224–236.

Belsky

Bakermans-Kranenburg

M. J.

van IJzendoorn

M. H.

(2007). For better and worse: Differential susceptibility to environmental influences. Current Directions in Psychological Science, 16, 300–304.

Ben-Amitay

Kimchi

Wolmer

Toren

(2016) Psychophysiological reactivity in child abuse. Journal of Child Sexual Abuse, 25, 185–200.

Boyce

W. T.

Ellis

B. J.

(2005). Biological sensitivity to context: I. An evolutionary-developmental theory of the origins of and functions of stress reactivity. Developmental Psychopathology, 17, 271–301.

Brown

E. J.

(2003). Child physical abuse: Risk for psychopathology and efficacy of interventions. Current Psychiatry Reports, 5, 87–94.

Busso

D. S.

McLaughlin

K. A.

Sheridan

M. A.

(2017). Dimensions of adversity, physiological reactivity and externalising psychopathology in adolescence: Deprivation and threat. Psychosomatic Medicine, 79, 162–171.

Cacioppo

J. T.

Uchino

B. N.

Bernston

G. G.

(1994). Individual differences in the autonomic origins of heart rate reactivity: The psychometrics of respiratory sinus arrhythmia and preejection period. Psychophysiology, 31, 412–419.

Carrey

N. J.

Butter

H. J.

Persinger

M. A.

Bialik

R. J.

(1995). Physiological and cognitive correlates of child abuse. Journal of the American Academy of Child and Adolescent Psychiatry, 34, 1067–1075.

10.

Chen

L. P.

Murad

M. H.

Paras

M. L.

Colbenson

K. M.

Sattler

A. L.

Goranson

E. N.

… Zirakzadeh

(2010). Sexual abuse and lifetime diagnosis of psychiatric disorders: Systematic review and meta-analysis. Mayo Clinic Proceedings, 85, 618–629.

11.

Cicchetti

(2010) Resilience under conditions of extreme stress: A multilevel perspective. World Psychiatry, 9, 145–154.

12.

Cicchetti

Toth

S. L.

(1995). A developmental psychopathology perspective on child abuse and neglect. Journal of the American Academy of Child and Adolescent Psychology, 34, 541–565.

13.

Cipriano

E. A.

Skowron

E. A.

Gatzke-Kopp

L. M.

(2011). Preschool children’s cardiac reactivity moderates relations between exposure to family violence and emotional adjustment. Child Maltreatment, 16, 205–215.

14.

Crowell

S. E.

Beuchaine

T. P.

Gatze-Kopp

Sylvers

Mead

Chipman-Chacon

(2006). Autonomic correlates of attention-deficit/hyperactivity-disorder and oppositional defiant disorder in preschool children. Journal of Abnormal Psychology, 115, 174–178.

15.

Dawson

M. E.

Schell

A. M.

Filion

D. L.

(2000). The electrodermal system. Handbook of Physiology, 2, 200–221.

16.

De Kloet

E. R.

Joels

Holsboer

(2005). Stress and the brain: From adaptation to disease. Nature Reviews Neuroscience, 6, 463–475.

17.

de Rooij

S. R.

(2013). Blunted cardiovascular and cortisol reactivity to acute psychological stress: A summary of results from the Dutch Famine Birth Cohort Study. International Journal of Psychophysiology, 90, 21–27.

18.

Del Giudice

Ellis

B. J.

Shirtcliff

E. A.

(2011). The adaptive calibration model of stress responsivity. Neuroscience and Biobehavioral Reviews, 35, 1562–1592.

19.

Ellis

B. J.

Essex

M. J.

Boyce

W. T.

(2005). Biological sensitivity to context: II. Empirical explorations of an evolutionary-developmental theory. Development and Psychopathology, 17, 303–328.

20.

El Sheikh

(2005). The role of emotional responses and physiological reactivity in the marital conflict-child functioning link. Journal of Child Psychology and Psychiatry, 46, 1191–1199.

21.

El Sheikh

Erath

S. A.

(2011). Family conflict, autonomic nervous system functioning, and child adaptation: State of the science and future directions. Development and Psychopathology, 23, 703–721.

22.

Fabes

R. A.

Eisenberg

Karbon

Troyer

Switzer

(1994). The relations of children’s emotional regulation to their vicarious emotional responses and comforting behaviours. Child Development, 65, 1678–1693.

23.

Ford

J. D.

Fraleigh

L. A.

Albert

D. B.

Conor

D. F.

(2010). Child abuse and autonomic nervous system hyporesponsivity among psychiatrically impaired children. Child Abuse and Neglect, 34, 507–515.

24.

Graziano

Derefinko

(2013). Cardiac vagal control and children’s adaptive functioning: A meta-analysis. Biological Psychology, 94, 22–37.

25.

Gunnar

M. R.

Frenn

Wewerka

S. S.

Van Ryzin

M. J.

(2009). Moderate versus severe early life stress: Associations with stress reactivity and regulation in 10–12 year old children. Psychoneuroendocrinology, 34, 62–75.

26.

Gunnar

M. R.

Quevedo

(2007). The neurobiology of stress and development. Annual Review of Psychology, 58, 145–173.

27.

Hayes

S. C.

Wilson

K. G.

Gifford

E. V.

Follette

V. M.

Strosahl

(1996). Experiential avoidance and behavioural disorders: A functional dimensional approach to diagnosis and treatment. Journal of Consulting and Clinical Psychology, 64, 1152–1168.

28.

Heleniak

McLaughlin

K. A.

Ormel

Riese

(2016). Cardiovascular reactivity as a mechanism linking child trauma to adolescent psychopathology. Biological Psychology, 120, 108–119.

29.

Hill

S. E.

Blechfield

Brunstetter

R. D.

Herbert

J. E.

Steckler

(1989). Cognitive and physiological responsiveness of abused children. Journal of the American Academy of Child and Adolescent Psychiatry, 28, 219–224.

30.

Hunter

A. L.

Minnis

Wilson

(2011). Altered stress responses in children exposed to early adversity: A systematic review of salivary cortisol studies. Stress, 14, 614–626.

31.

Jenness

J. L.

Bryant Miller

Rosen

M. L.

McLaughlin

K. A.

(2018). Extinction learning as a potential mechanism linking high vagal tone with lower PTSD symptoms among abused youth. Journal of Abnormal Child Psychology, 47, 1–12.

32.

Kendall-Tackett

K. A.

Williams

L. M.

Finkelhor

(1993). Impact of sexual abuse on children: A review and synthesis of recent empirical studies. Psychological Bulletin, 113, 164–180.

33.

Koopman

Carrion

Butler

L. D.

Sudhakar

Palmer

Steiner

(2004). Relationships of dissociation and childhood abuse and neglect with heart rate in delinquent adolescents. Journal of Traumatic Stress, 17, 47–54.

34.

Leitzke

B. T.

Hilt

L. M.

Pollak

S. D.

(2015). Maltreated youth display a blunted blood pressure response to an acute interpersonal stressor. Journal of Clinical Child and Adolescent Psychology, 44, 305–313.

35.

Lunkenheimer

Busuito

Brown

K. M.

Panlilio

Skowron

E. A.

(2019). The interpersonal neurobiology of child maltreatment: Parasympathetic substrates of interactive repair in maltreating and nonmaltreating mother-child dyads. Child Maltreatment. doi:10.1177/1077559518824058

36.

MacMillan

H. L.

Georgiades

Duku

E. K.

Shea

Steiner

Niec

… Schmidt

L. A.

(2009). Cortisol response to stress in female youths exposed to childhood maltreatment: Results of the youth mood project. Biological Psychiatry, 66, 62–68.

37.

Matthys

van Goozen

S. H. M.

Snoek

van Engeland

(2004). Response perseveration and sensitivity to reward and punishment in boys with oppositional defiant disorder. European Child and Adolescent Psychiatry, 13, 362–364.

38.

McLaughlin

K. A.

Alves

Sheridan

M. A.

(2014). Vagal regulation and internalising psychopathology among adolescents exposed to childhood adversity. Developmental Psychobiology, 56, 1036–1051.

39.

McLaughlin

K. A.

Sheridan

M. A.

Alves

Berry Mendes

(2014). Child maltreatment and autonomic nervous system reactivity: Identifying dysregulated stress reactivity patterns by using the biopsychosocial model of challenge and threat. Psychosomatic Medicine, 76, 538–546.

40.

McLaughlin

K. A.

Sheridan

M. A.

Gold

A. L.

Duys

Lambert

H. K.

Peverill

… Pine

D. S.

(2016). Maltreatment exposure, brain structure, and fear conditioning in children and adolescents. Neuropsychopharmacology: Official publication of the American College of Neuropsychopharmacology, 41, 1956–1964.

41.

McLaughlin

K. A.

Sheridan

M. A.

Tibu

Fox

N. A.

Zeanah

C. H.

Nelson

C. A.

(2015). Causal effects of the early caregiving environment on development of stress response systems in children. Proceedings of the National Academy of Sciences of the United States of America, 112, 5637–5642.

42.

Miscovic

Schmidt

L. A.

Georgiades

Boyle

MacMillan

H. L.

(2009). Stability of resting frontal electroencephalogram (EEG) asymmetry and cardiac vagal tone in adolescent females exposed to child maltreatment. Developmental Psychobiology, 51, 474–487.

43.

Mulvihill

(2005). The health impact of childhood trauma: An interdisciplinary review, 1997–2003. Issues in Comprehensive Pediatric Nursing, 28, 115–137.

44.

Nemeroff

C. B.

(2016). Paradise lost: The neurobiological and clinical consequences of child abuse and neglect. Neuron, 89, 892–909.

45.

Oosterman

De Schipper

J. C.

Fisher

Dozier

Schuengel

(2010). Autonomic reactivity in relation to attachment and early adversity among foster children. Development and Psychopathology, 22, 109–118.

46.

Paolucci

E. O.

Genius

M. L.

Violato

(2001). A meta-analysis of the published research on the effects of child sexual abuse. Journal of Psychology, 135, 17–36.

47.

Parker

K. J.

Buckmaster

C. L.

Sundlass

Schatzberg

A. F.

Lyons

D. M.

(2006). Maternal mediation, stress inoculation and the development of neuroendocrine stress resistance in primates. Proceedings of the National Academy of Sciences of the United States of America, 103, 3000–3005.

48.

Phillips

A. C.

Ginty

A. T.

Hughes

B. M.

(2013). The other side of the coin: Blunted cardiovascular and cortisol reactivity are associated with negative health outcomes. International Journal of Psychophysiology, 90, 1–7.

49.

Pollak

S. D.

Vardi

PutzerBechner

A. M.

Curtin

J. J.

(2005). Physically abused children’s regulation of attention in response to hostility. Child Development, 76, 968–977.

50.

Porges

S. W.

(2004). Social engagement and attachment: A phylogenetic perspective. Annals of the New York Academy of Sciences, 1008, 31–47.

51.

Porges

S. W.

(2007). The polyvagal perspective. Biological Psychology, 74, 116–143.

52.

Quas

J. A.

Hong

Alkon

Boyce

W. T.

(2000). Dissociations between psychobiologic reactivity and emotional expression in children. Developmental Psychobiology, 37, 153–175.

53.

Richter

Gendolla

G. H. E.

(2009). The heart contracts to reward: Monetary incentives and preejection period. Psychophysiology, 46, 451–457.

54.

Saltzman

K. M.

Holden

G. W.

Holahan

C. J.

(2005). The psychobiology of children exposed to marital violence. Journal of Clinical Child and Adolescent Psychology, 34, 129–139.

55.

Schachinger

Weinbacher

Kiss

Ritz

Langewitz

(2001). Cardiovascular indices of peripheral and central sympathetic activation. Psychosomatic Medicine, 63, 788–796.

56.

Shenk

C. E.

Noll

J. G.

Putnam

F. W.

Trickett

P. K.

(2010). A prospective examination of the role of childhood sexual abuse and physiological asymmetry in the development of psychopathology. Child Abuse & Neglect, 34, 752–761.

57.

Shenk

C. E.

Putnam

F. W.

Noll

J. G.

(2012). Experiential avoidance and the relationship between child maltreatment and PTSD symptoms: Preliminary evidence. Child Abuse & Neglect, 36, 118–126.

58.

Shenk

C. E.

Putnam

F. W.

Rausch

J. R.

Peugh

J. L.

Noll

J. G.

(2014). A longitudinal study of several potential mediators of the relationship between child maltreatment and post-traumatic stress disorder. Development and Psychopathology, 26, 81–91.

59.

Snoek

Van Goozen

S. H. M.

Matthys

Buitelaar

J. K.

(2004). Stress responsivity in children with externalising behaviour disorders. Development and Psychopathology, 16, 389–406.

60.

Staufenbiel

S. M.

Penninx

B. W. J. H.

Spijker

A. T.

Elzinga

B. M.

van Rossum

E. F. C.

(2013). Hair cortisol, stress exposure, and mental health in humans: A systematic review. Psychoneuroendocrinology, 38, 1220–1235.

61.

Van Goozen

S. H. M.

Matthys

Cohen-Kettenis

P. T.

Buitelaar

J. K.

Van Engeland

(2000). Hypothalamic-pituitary-adrenal axis and autonomic nervous system activity in disruptive children and matched controls. Journal of the American Academy of Child and Adolescent Psychiatry, 39, 1438–1445.