Early Life Adversity and Males: Biology, Behavior, and Implications for Fathers’ Parenting

Abstract

Fathers have an important and unique influence on child development, but influences on fathers’ parenting have been vastly understudied in the scientific literature. In particular, very little empirical research exists on the effects of early life adversity (ELA; e.g. childhood maltreatment, parental separation) on later parenting among fathers. In this review, we draw from both the human and non-human animal literature to examine the effects of ELA, specifically among males, in the following areas: 1) neurobiology and neurocognitive functioning, 2) hormones and hormone receptors, 3) gene-environment interactions and epigenetics, and 4) behavior and development. Based on these findings, we present a conceptual model to describe the biological and behavioral pathways through which exposure to ELA may influence parenting among males, with a goal of guiding future research and intervention development in this area. Empirical studies are needed to improve understanding of the relationship between ELA and father’s parenting, inform the development of paternal and biparental interventions, and prevent intergenerational transmission of ELA.

1. Introduction

Supportive and responsive parenting by mothers, fathers or other caregivers is necessary for optimal child development. Critical skills for supportive parenting include stress tolerance, mentalization and reflective functioning (i.e., an ability to reflect on the thoughts, feelings, and intentions of oneself and others), and the development of secure parent-child attachment relationships (Britto et al., 2017; Slade, 2005). A vast body of research conducted with mothers demonstrates that early life adversity (ELA), including experiences such as childhood maltreatment, poverty, and prolonged or repeated parental separation, may impair these critical parenting skills and contribute to intergenerational transmission of adversity (Lomanowska et al., 2017; Savage et al., 2019). However, despite evidence that the biological and behavioral effects of ELA may differ for males and females (e.g., Desantis et al., 2011; Everaerd et al., 2016; Logan-Greene & Semanchin Jones, 2015), the effects of ELA on fathers’ parenting have not been well studied (Christie et al., 2017; Lomanowska et al., 2017). There is also a concerning dearth of interventions targeted specifically towards fathers, particularly those at risk for parenting challenges due to ELA history (Pruett et al., 2019; Stover et al., 2020). Thus, the purpose of this review is to examine the effects of ELA on male biology and behavior, explore implications for parenting among fathers with a history of ELA, and develop a conceptual model to help guide future research and intervention development in this critical area.

1.1. The Importance of Fathers’ Parenting

Fathers influence children’s development both directly through caregiving behaviors and indirectly by influencing the quality of other caregiver-child interactions (Cabrera et al., 2018; Sarkadi et al., 2008). Evidence suggests that fathers often interact with their children in ways that differ from mothers, such as more frequently engaging in stimulatory, tactile, and exploratory behaviors (Feldman, 2003; Rajhans et al., 2019). Other research indicates that fathers’ parenting is similar to mothers’ parenting, is influenced by family context (e.g., single fathers, same sex parents), and may even buffer the effects of negative environmental stimuli, such as maternal depression, on child outcomes (Cabrera et al., 2014; Cabrera et al., 2018; Lewin et al., 2014). Together, these findings indicate that fathers’ parenting is an important complement to parenting provided by mothers and other caregivers and may uniquely contribute to children’s development (Cabrera et al., 2014; Feldman, 2003; Rajhans et al., 2019).

Past research with fathers has largely focused on examining the effects of fathers’ parenting on child development. However, less is known about the complex factors that influence fathers’ parenting behaviors themselves, especially factors that occur prior to the transition to parenthood. In their expanded model of influences on fathering, Cabrera and colleagues (2014) posit that fathers’ history, including biological and rearing history, influences fathers’ personal characteristics, which in turn influences fathers’ parenting behaviors. ELA is one such component of paternal history that may influence fathers’ parenting, as studies with both human and nonhuman animals demonstrate that ELA may affect later biology and behavior (McCrory et al., 2017). For example, prospective, empirical studies in animals indicate that ELA in the form of maternal or paternal deprivation results in more aggressive and less social male offspring with altered neurobiological functioning (Maccari et al., 2014; Suomi, 1991; Veenema, 2009). However, very few empirical studies have examined relationships between ELA and parenting among males, and thus the effects of ELA on fathers’ parenting remain poorly understood.

1.2. Current Knowledge on the Relationship between ELA and Fathers’ Parenting

Of the few studies that have examined fathers who experienced ELA, results from human studies suggest that ELA does influence fathers’ parenting, and in ways that may differ from influences on mothers’ parenting. For example, Skjothaug and colleagues (2014) found that retrospectively reported ELA as measured by the Adverse Childhood Experiences scale was associated with fathers’ depression and anxiety during their partners’ pregnancy (Skjothaug et al., 2015), and that this depression and anxiety mediated a relationship between ELA and parental stress at 6 months postpartum (Skjothaug et al., 2018). In a prospective study, Szepsenwol and colleagues (2015) found that unpredictability during the first four years of life (i.e., frequent changes in parental employment status, cohabitation status, and residence) predicted more negative parenting orientations (e.g. hostility, less emotional connectedness) in fathers, but not mothers. However, in a separate study, retrospectively-reported unpredictability in early life was associated with more negative co-parental behaviors (e.g. less coordination with a partner) among mothers, but not fathers (Szepsenwol, 2020).

Some evidence also suggests that fathers’ parenting behaviors are transmitted across generations. For example, Capaldi and colleagues (2008) found that fathers who experienced poor discipline (e.g., punishment dependent on caregiver’s mood) during childhood were more likely to engage in poor and harsh disciplinary practices (e.g., spanking) with their toddlers (Capaldi et al., 2008). In contrast, males who reported close, supportive relationships with their own fathers were more likely to report positive relational schemas, which in turn predicted involvement with their young children (Brown et al., 2018). In a study with new and expectant fathers, Alyousefi-van Dijk et al. (2020) found that higher childhood maltreatment history was associated with more use of excessive handgrip force in response to infant cry sounds. Handgrip is a measure of one’s ability to modulate behavior in response to stressor, and excessive handgrip force has been associated with insecure attachment history. Though limited, these studies demonstrate that ELA may have important influences on fathers’ parenting that differ from effects on mothers’ parenting. Therefore, to gain a robust understanding of the effects of ELA on fathers’ parenting, research must be conducted with fathers specifically.

2. The Current Review

The transition to fatherhood is associated with neurodevelopmental and hormonal changes in males (see Tables 1 and 2), and the targets of these changes support fathers’ parenting behavior. A growing body of literature also suggests parenting is influenced by individual genetic, behavioral, and developmental differences (Beaver & Belsky, 2012; Cabrera et al., 2014; Klahr & Burt, 2014). Notably, ELA is also associated with altered neurobiological, hormonal, epigenetic, and behavioral outcomes, and further, many studies note differences in the effects of ELA by sex or gender (e.g., Desantis et al., 2011). Therefore, in order to understand how ELA may influence parenting among fathers, we review evidence for the effects of ELA, specifically on males, in each of the following areas: 1) neurobiology and neurocognitive functioning, 2) hormones, 3) gene-environment interactions and epigenetics, and 4) behavior and development. We then discuss possible implications for parenting among fathers who experienced ELA and directions for future research in this area. Based on the findings of this review and grounded in theoretical models of parenting (Belsky, 1984; Cabrera et al., 2014), we present a conceptual model to describe the potential biological and behavioral connections between ELA and fathers’ parenting (Figure 1). As research in this area is still emerging, our conceptual model is intended as a springboard to guide future research on the relationship between ELA and fathers’ parenting, and to help identify potential targets for intervention with fathers who experienced ELA.

Table 1.

Summary of key brain networks implicated in fathers’ parenting (Abraham, Hendler, Zagoory-Sharon, & Feldman, 2016; Feldman, Braun, & Champagne, 2019).

Network	Role	Key Brain Regions
Reward-motivational networks (also referred to as core-limbic network)	Associated with paternal emotional experience and heightened feelings of reward and social bonding towards infants. Circuitry is conserved across species and important for both maternal and paternal care.	Mid-brain dopaminergic network (e.g. Nucleus Accumbens, Ventral Tegmental area [VTA]) Globus Pallidus Amygdala Putamen Medial thalamus Hypothalamus Lateral Septum Bed nucleus of the stria terminalis (BNST)
Empathy and embodied simulation networks	Supports empathy towards the infant’s affective state. Coupling of perceptual-motor regions (e.g. the IFG and IPL) enable simulation of the infant’s actions. Activation of these networks underlie sensitive caregiving and ground the infant’s experiences to the present moment.	Anterior Insula (AI) Medial/Anterior cingulate cortex (ACC) Inferior Frontal Gyrus (IFG) Inferior Parietal Lobe (IPL) Superior Temporal Sulcus (STS)
Mentalizing and emotion regulation networks	Support fathers’ understanding of their infant’s actions and intentions, whilst also supporting executive functions (e.g. multitasking, planning) and emotion regulation. Important to how fathers behave toward their infants.	Dorsal medial pre-frontal cortex (dmPFC) Ventro-medial PFC (vmPFC) Temporo-Parietal Junction Temporal pole Superior Temporal Sulcus (STS) Frontopolar cortex

Note. Interpretation of these networks requires caution, as each area often has multiple functions. The regions listed have been identified based on prior fMRI research as active in fathers in response to infant stimuli. The networks are defined based on meta-analyses of neuroimaging studies indicating common co-activation (Abrahams et al., 2016).

Table 2.

Hormones Associated with Fathers’ Parenting and the Transition to Fatherhood

Hormone	Role in Fathers’ Parenting	Changes in Fatherhood	References
Estradiol (E2)	Converted from T to activate parental behavior in some species	Variability in fathers according to species and measurement	Saltzman & Ziegler, 2014; Bales & Saltzman, 2016; Storey & Ziegler, 2016; Storey & Walsh, 2011; Gettler et al., 2018
Testosterone (T)	Organizing brain circuitry related to paternal behavior; activating adult/parental behavior (via conversion to E2) in some species	Some increased circulating levels during mate’s gestation, but decreased levels in fathers following birth	Bales & Jarcho, 2013; Saltzman & Ziegler, 2014; Bales & Saltzman, 2016; Storey & Ziegler, 2016; Storey & Walsh, 2011; Gettler, et al., 2011a, b; Rilling & Mascaro, 2017
Oxytocin (OT)	Associated with food-sharing tolerance, infant interactions, and parent-infant affect synchrony	Inconsistent findings in brain signaling and circulating levels in fathers across animal species	Saltzman & Ziegler, 2014; Bales & Saltzman, 2016; Storey & Ziegler, 2016; Rilling & Mascaro, 2017
Vasopressin (AVP)	Associated with a wide range of male reproductive behaviors & activities	Inconsistent findings in brain signaling and circulating levels in fathers across animal species	Saltzman & Ziegler, 2014; Bales & Saltzman, 2016; Storey & Ziegler, 2016
Glucocorticoids (GC)	Facilitates paternal behavior in new fathers; associated with infant carrying & responsiveness to infant cries	Increased circulating levels during mate’s gestation and drop postpartum	Bales & Jarcho, 2013; Saltzman & Ziegler, 2014; Bales & Saltzman, 2016; Storey & Ziegler, 2016; Gettler et al., 2011b.

Note: This is a general overview only which does not take into account interactions between individual hormones. Readers are directed to the references for more detailed information.

Figure 1. Potential Biological and Behavioral Influences on Parenting among Males who Experienced Early Life Adversity.

Note: The effects of early life adversity vary based on individual circumstances, including the type, timing, and severity of adversity experienced and the presence of protective factors. Fathers are also situated within a dynamic, transactional context which is influenced by the current family-, community-, and societal-level environment.

2.1. Defining Early Life Adversity

ELA is a broad term used to describe prenatal and postnatal experiences that may have a harmful effect on child health or development. For the purposes of this review, we focus on postnatal experiences of ELA, including but not limited to abuse, neglect, childhood poverty, parental separation, or exposure to parental mental illness. When possible, we describe the specific ELA exposure measured, as the effects of ELA may vary based on adversity type, timing, chronicity, or severity (McLaughlin et al., 2014). We also acknowledge the inherent challenges and limitations of ELA research, including incongruence between retrospective and prospective measurements of ELA (Baldwin et al., 2019) and the potential buffering effects of childhood protective factors, which may be infrequently measured (Yule et al., 2019). While it is beyond the scope of this review to explore the influence of these factors in depth, we include these concepts in our conceptual model to highlight the importance of these considerations in future research.

2.2. Defining Fathers

We define fathers broadly in recognition that the father role varies across cultures and contexts and may extend to other male caregivers beyond the biological father, such as adoptive fathers, uncles, grandfathers, step-fathers, and others (Cabrera et al., 2000). Thus, while most past research on fathers has focused on biological fathers, our conceptual model may apply to any male that serves a father role.

2.3. The Contribution of Animal Studies

Within each section of this review, we focus primarily on evidence from human studies; however, we also highlight animal studies that have provided foundational knowledge or have helped fill in gaps in knowledge that are difficult, if not impossible, to study in human subjects such as prospective studies in which ELA can be experimentally assigned. The majority of such studies we discuss occur in mammals that exhibit fathering behavior and/or paternal care along with adult pair-bonding, specifically rodents and nonhuman primates, owing to their greater similarity to humans in genetics, physiology, behavior, and life histories as compared to bird or fish species (Kentner et al., 2010). Such animal models provide unique insight into evolutionarily conserved biological pathways and allow a level of control and determination of causality that is usually not possible in studies with humans. However, it is important to note that no animal model provides a perfect translation to human fathers, and certain methods are better suited for answering research questions in humans, such as neuroimaging for understanding neural function at a systems level. A comprehensive review of fathering behavior in nonhuman animals is outside the scope of this paper, especially as others have so elegantly provided such reviews (Bales & Jarcho, 2013; Bales & Saltzman, 2016; Fernandez-Duque et al., 2009; Kentner et al., 2010; Rilling & Mascaro, 2017; Rogers & Bales, 2019; Saltzman & Ziegler, 2014; see also Tables 2–6). Furthermore, we must note that while we cite and discuss research in rats examining the effects of ELA on neurobiology, hormone functioning, and behavior of males to highlight direction for examination in humans, male rats do not exhibit fathering behaviors and so this particular animal model is not adequate for studying the effects of ELA on male parenting behavior. Rather, for these models we focus on animals that do exhibit male parenting behavior, including other rodents and some nonhuman primates. Nonetheless, the value of animal models in adding to our basic knowledge of brain, genetic, hormonal, and other outcomes of ELA cannot be overstated.

Table 6.

Example Evidence of the Effects of Early Life Adversity on Males’ Behavior & Development

Citation	Study Design	Species	N	Early Life Adversity (ELA)			Key Findings
				Type	Timing	Measure
Externalizing Behaviors & Mental Health(Cyr et al., 2010; Godinet et al., 2014; Leban & Gibson, 2020; Maccari et al., 2014; Seff & Stark, 2019; Suomi, 1991, 1997; Terrell et al., 2019; Veenema, 2009)
Suomi, 1991, 1997; Veenema, 2009;	Review	Rodents and non-human primates	Varies	Repeated maternal separations or maternal absence	Varies	N/A	ELA induces impulsive, excessive and inappropriate aggression toward social group members among males during adolescence and adulthood
Maccari et al., 2014	Review	Rodents	Varies	Maternal separation or maternal deprivation	Varies	N/A	Male rodent offspring exposed to ELA show increased risk-taking behaviors in a rat model of stress-related diseases across the lifespan
Terrell et al., 2019	P, L	Humans	1388	Index of nine adversities (e.g. maternal substance use, poverty)	Birth to age 3	PR, O	ELA associated with increased aggressive behavior at age 12 to 14 years among males, but not females
Cyr et al., 2017	CS	Humans	1400	Victimization (e.g. crime, maltreatment, exposure to violence)	Birth to current age (12 to 17 years)	RSR	Sexual victimization predicted anger and aggressive behavior in adolescent females. History of physical assault and witnessing violence predicted anger in adolescent males.
Godinet, et al., 2014	P, L	Humans	484	Maltreatment	Birth to age 4 years	CPS	Males with allegations of maltreatment had more externalizing behavior problems than controls at age 4, but effects decreased over time and groups were not significantly different at age 10. Among females, effects increased over time, with no group differences at age 4 but more behavioral problems in the maltreatment allegation group noted by age 12.
Leban & Gibson, 2020	P, L	Humans	1911	Adverse childhood experiences (ACEs)	Birth to current age (Mean 12 years)	SR	Among adolescents, childhood ACEs were associated with increased delinquency among males only and with substance use among females only
Seff & Stark, 2019	CS	Humans	2670 (Kenya sample)	Emotional violence	Birth to current age (Mean 18 years)	RSR	Emotional violence associated with increased risk for suicidal ideation among males and females in Haiti, Kenya and Tanzania. In Kenya only, this association was statistically larger among males than females.
Attachment (Bales & Saltzman, 2016; Beeghly et al., 2017; Carlson et al., 1989; Kobulsky et al., 2016; Lyons- Ruth et al., 1999; Murray, 1992; Rogers & Bales, 2019)
Bales & Saltzman, 2016; Rogers & Bales, 2019	Review	Rodents	Varies	Paternal deprivation via removal of father	Varies	N/A	Male offspring exposed to father removal show deficits in pair-bonding as adults
Carlson et al., 1989	CS	Humans	43 dyads	Maltreatment	Birth to 16 months	CPS	Maltreated children were more likely to have disorganized attachment than non-maltreated children. Males were more likely to have disorganized attachment than females, regardless of maltreatment status.
Lyons-Ruth et al., 1999	CS	Humans	65 dyads	Maltreatment, atypical maternal behavior	Birth to 18 months	CPS, O	Among mothers with history of maltreatment, disorganized attachment was more common among male infants than female infants
Beeghly et al., 2017,	L, P	Humans	182 dyads	Maternal depression	Birth to 18 months	PR	Rapid increase in maternal depressive symptoms was associated with lower attachment security at 18 months, but only for male children.
Murray, 1992	L, P	Humans	111	Maternal depression	Birth to 18 months	PR	Among children exposed to maternal depression, males were 3.6 times more likely than females to have insecure attachment
Kobulsky et al, 2016	L	Humans	297 dyads	Exposure to violence (e.g. domestic violence, bullying community violence)	Birth to age 10 years	SR	ELA predicted tobacco and substance use at age 12 years for both males and females. In females only, this relationship was mediated by depression and anxiety symptoms at age 11. In males only, adoptive status was a risk factor for alcohol use and higher home quality was protective against substance use.
Substance Misuse (Hudson et al., 2017; Huggins et al., 2012)
Huggins et al., 2012	E, P	Rhesus monkeys	18	Maternal separation in infancy (via nursery rearing)	Up to 7 months postnatal	N/A	As young adults, ELA monkeys consumed more alcohol in a free-drinking paradigm than controls, and also exhibited lower serotonin functioning in the brain
Hudson et al., 2017	CS	Humans	301	Childhood Sexual Abuse	Birth to 18 years (mean age 15.9 years)	SR	Relationship between childhood sexual abuse and problem drinking was fully mediated by anger and anxiety symptoms among females, and partially mediated by anger symptoms among males.

Note: Study Design: E=Experimental, P=Prospective, L=Longitudinal, CS=Cross-sectional

ELA Measure: CPS=Child Protective Services records; O=observation PR=Parent report; RSR=Retrospective self-report; SR=Self report

3. ELA and Neurobiology and Neurocognitive Functioning

3.1. Neurobiology

Animal models provide insight into the neurobiological effects of ELA that cannot be easily tested in humans, including neural mechanisms and cellular and molecular neurobiology (Table 3; see Bales & Saltzman, 2016; Feldman et al., 2019 for comprehensive reviews). While very few animal studies have focused on outcomes among males specifically, some studies in rats have shown that adult male offspring of low-caring mothers have reduced synaptic plasticity when tested in non-stressed conditions, but increased plasticity in stressed conditions, when compared to high-caring mothers (Bagot et al., 2009; Champagne, 2008). These findings suggest that early adversity contributes to later ability to learn and adapt under conditions of varying stress levels. In studies of biparental mandarin voles, paternally deprived males had enhanced expression of dopamine receptors in the nucleus accumbens, an area associated with motivational and reward processing (Yu et al., 2012). Studies of a different rodent lineage, Octodon degus, demonstrate that paternal deprivation results in differential innervation in stress-responsive systems (reviewed in Bales & Saltzman, 2016). Because of evolutionarily conserved brain structures in human and nonhuman mammals, the foundational knowledge provided by these and other animal studies elucidates the neurobiology underlying similar processes in humans (see Feldman et al., 2019; Bales & Saltzman, 2016). Collectively, the animal literature indicates that ELA affects the development of the stress response and motivation/reward processing systems in male offspring.

Table 3.

Example Evidence of the Effects of Early Life Adversity on Males’ Neurobiology and Neurocognitive Functioning

Citation	Study Design	Species	N	Early Life Adversity (ELA)			Key Findings
				Type	Timing	Measure
Neurobiology
Bagot et al., 2009	E, P	Rats	30	Low-caring mothers	Up to 21 days (weaning)	N/A	Adult male offspring of low-caring mothers have reduced hippocampal dendritic complexity and improved dentate plasticity during stress when compared to offspring of high-caring mothers
Champagne, et al., 2008	E, P	Rats	13	Low-caring mothers	Up to 21 days (weaning)	N/A	Adult male offspring of low-caring mothers have reduced synaptic plasticity when tested in non-stressed conditions, but increased plasticity in stressed conditions, when compared to offspring of high-caring mothers
Wang et al., 2012	E, P	Mandarin voles	12–18	Paternal deprivation	Up to 23 days (weaning)	N/A	Paternally deprived males had differential AVP immunoreactivity in various brain regions compared to controls (biparentally-raised males)
Yu et al., 2012	E, P	Mandarin voles	10–18	Paternal deprivation	Up to 23 days (weaning)	N/A	Paternally deprived males had enhanced expression of dopamine receptors in the nucleus accumbens
Braun et al., 2013	E, P	Octodon degus	24	Paternal deprivation	Up to 21 (weaning) or 45 days (puberty)	N/A	Paternal deprivation results in differential innervation in stress-responsive systems compared to controls
Neural Structure
Jensen et al., 2015	P, L	Humans	494	Cumulative index of 37 family adversities	Birth to age 6	PR	ELA associated with lower grey matter volumes in right and higher grey matter volumes in right precuneus at age 18 to 21 years, controlling for other adversities.
Roth et al., 2018	CS	Humans	138	Neglect	Up to current age (9 to 15 years)	RSR	Larger right amygdala volumes positively correlated with history of neglect in males but not in females. Right amygdala volume mediated association between neglect and anxiety symptoms in males.
Ugwu, et al., 2015	CS	Humans	92	Maltreatment	Birth to age 16	RSR	Males reporting ELA history, but not females, had altered structural connectivity of white matter tracts (e.g. cingulum and uncinate fasciculus) connecting limbic and prefrontal brain regions
Neurocognitive Functioning
Crozier et al., 2014	CS	Humans	74	Maltreatment	Birth to current age (8 to 16 years	CPS	Maltreated males had increased activity in the visual cortex and lingual gyrus and decreased activity in the cingulate cortex in response to fearful faces, compared to non-maltreated males. Relationship not found in females.
Elton et al. 2014	CS	Humans	40	Maltreatment	Birth to age 16	RSR	Maltreatment dose-related decrease in dorsal ACC activity, modulated by left inferior frontal cortex, predicted greater response inhibition in adult females but poorer response inhibition in adult males.
LoPilato et al., 2020	L	Humans	605	Threat (e.g. abuse) and deprivation (e.g. poverty, neglect)	Birth to age 16	RSR	Among adolescents and young adults, heightened stress perception was associated with deprivation-based ELA in males and with threat-based ELA in females.
Nooner et al., 2018	CS	Humans	193	Maltreatment	Birth to current age (12 years)	CPS	Females with ELA history performed better than males with ELA history on a short-term verbal memory task

Note: ACC, Anterior Cingulate Cortex;

Study Design: E=Experimental, P=Prospective, L=Longitudinal, CS=Cross-sectional

ELA Measure: CPS=Child Protective Services records; PR=Parent report; RSR=Retrospective self-report Early Life Adversity and Males 54

3.2. Neural Structure

A small number of human studies have noted sex differences in the impact of ELA on neural structure (Table 3). For example, in a study of males who participated in the Avon Longitudinal Study of Parents and Children, ELA in the first 6 years of life (a cumulative index of 37 adversities, including interpersonal loss, family instability, and child maltreatment) was associated with lower grey matter volumes in the anterior cingulate cortex (ACC) and higher grey matter volumes in precuneus at age 18 to 21 years. These areas are implicated in the empathy and embodied simulation networks important in human fathering (Feldman et al., 2019; Jensen et al., 2015). In a study of adolescents, self-reported childhood neglect was positively correlated with right amygdala volumes in males but not in females, and amygdala volume mediated an association between neglect and anxiety symptoms among males (Roth et al., 2018). In a study of adults, Ugwu and colleagues (2015) found altered structural connectivity of white matter tracts (e.g., cingulum and uncinate fasciculus) connecting limbic and prefrontal brain regions in males but not females with histories of childhood maltreatment, suggesting altered structural connectivity between regions associated with emotion regulation. Taken together, these findings indicate ELA in males is associated with changes in neural structure in regions involved in empathy and emotion regulation.

3.3. Neurocognitive Functioning

ELA is associated with greater reactivity towards negative and threatening emotional stimuli (da Silva Ferreira et al., 2014; McCrory et al., 2017), and emerging evidence suggests that underlying neural functioning differs by sex. For example, in a neuroimaging study of school-aged children with and without histories of maltreatment, Crozier et al. (2014) found that maltreated males dedicate significant neural resources to the processing of fear and are more vulnerable to distraction by fearful stimuli than females (see Table 3). Sex differences in stress perception and reactivity may also be influenced by adversity type (McLaughlin et al., 2014). For example, in a study of adolescents and young adults, history of deprivation-based adversity (e.g., poverty, neglect, parental absence) was associated with heightened stress perception among males, while history of threat-based adversity (e.g., abuse, bullying) was associated with heightened stress perception in females (LoPilato et al., 2020).

ELA may also influence emotion regulation more broadly, as individuals with histories of childhood maltreatment demonstrate altered brain activity in regulatory regions such as the ACC lateral prefrontal cortex, as well as altered functional connectivity between frontal and subcortical brain regions such as the medial prefrontal cortex and the amygdala (Gee et al., 2015; McCrory et al., 2017). These functional networks are particularly important in emotion regulation, and atypical activation in these regions may indicate latent vulnerability for psychopathology in individuals who have experienced ELA (McCrory et al., 2017). Further, emotion regulation strategies also differ by sex; compared with females, males are less likely to ruminate on emotional experiences and also to engage with coping strategies involving verbal expression of emotions (Johnson & Whisman, 2013; Tamres et al., 2002), which may be influenced by socialization (Berke et al., 2018). However, no functional neuroimaging studies have examined sex differences in emotion regulation among individuals who experienced ELA.

ELA is also known to impair executive functioning. Meta-analyses demonstrating that childhood maltreatment impairs working memory, attention, and speed of processing, with effects persisting into adulthood (Masson et al., 2015; McCrory et al., 2017; Nikulina & Widom, 2013). While few studies have examined sex differences, limited evidence suggests males may be more vulnerable to the harmful effects of ELA on executive functioning, including effects on verbal memory (Nooner et al., 2018) and inhibitory control (Elton et al., 2014).

4. Relationships between Males’ ELA and Hormones and Hormone Receptors

4.1. Testosterone

Testosterone (T) is an anabolic steroid hormone that has been associated with mating, caregiving behaviors, and aggression among males (Montoya et al., 2012; Rajhans et al., 2019). While research in human and animal models is limited, collectively, evidence suggests that ELA may affect later T functioning in males (Table 4), particularly in response to current stress. For example, Veenema and colleagues (2006) found that adult male rats exposed to maternal separation in the neonatal period exhibited reduced plasma T in response to stress compared with basal T levels. In human adolescents, Simmons et al (2015) found that males’ experience of maternal aggression at child age 12 moderated the relationship between cortisol and testosterone at age 15. In one longitudinal study of Filipino males, harshness or unpredictability during childhood was associated with elevated waking T in fathers; however, it is unclear how this may have affected parenting, as fathers’ parenting behaviors were not examined (Sarma et al., 2018).

Table 4.

Example Evidence of the Effects of Early Life Adversity on Males’ Hormones

Citation	Study Design	Species	N	Early Life Adversity (ELA)			Key Findings
				Type	Timing	Measure
Testosterone (T) (Pascual-Sagastizabal et al., 2014; Sarma et al., 2018; Siegeler et al., 2013; Simmons et al., 2015; Veenema et al., 2006; Zito et al., 2017)
Zito et al., 2017	E, P	Zebra finches	84	Nutritional stress	3–33 days post hatching	N/A	Male chicks had higher plasma T than controls at day 60
Siegeler et al., 2013	E, P	Wild cavies	24	Maternal exposure to social instability	Up to 20 days (weaning)	N/A	Male offspring had delayed increases in plasma T around adolescence compared to controls, but no differences in adulthood
Veenema et al., 2006	E, P	Rats	15	Maternal separation	Up to day 14	N/A	As adults, males exposed to maternal separation exhibited reduced plasma T in response to stress compared with basal T levels
Sarma et al., 2018	L	Humans	417	Death of a sibling or paternal instability	Birth to age 11.5 years	PR	Higher elevated waking T observed in fathers who experienced sibling death and who had younger age at sexual debut. New fathers who experienced paternal instability and who had younger age at sexual debut had declines in waking T compared to other new fathers.
Pascual-Sagastizabal et al., 2014	CS	Humans	159	Parenting (including corporal punishment)	8 years old	PR	Among 8 year-olds, high T was associated with physical aggression, but only among boys with authoritarian mothers
Simmons et al., 2015	L	Humans	89	Maltreatment, maternal aggression, stressful life events	Birth to age 14	RSR, O	Relationship between cortisol and testosterone at age 15 was positive for males with maternal aggression observed at age 12
Oxytocin (OT) (Bales & Perkeybile, 2012; Lukas & Clutton-Brock, 2013; Mizuki & Fujiwara, 2015; Opacka-Juffry & Mohiyeddini, 2012; Seltzer et al., 2014)
Lukas et al., 2010	E, P	Rats	95	Maternal separation	Up to day 14	N/A	Male rats exposed to maternal separation exhibit reduced OT receptor expression in multiple brain regions across development compared to controls
Bales and Perkeybile, 2012	Review	Mice, rats, and prairie voles	varies	Single mothers or mothers showing reduced care	Varies by study and species	N/A	Rearing by single mothers or mothers showing reduced care associated with reduced OT receptor expression in many brain regions as adults
Opacka-Juffry & Mohiyeddini, 2012	CS	Humans	90	Stressful childhood experiences	Up to age 12	RSR	In adult males, lower plasma OT was associated with stressful experiences in childhood, but OT not associated with adolescent or recent stressful life events
Mizuki & Fujiwara, 2015	CS	Humans	80	Maltreatment	Birth to age 16 years	RSR	Childhood physical abuse associated with increased urinary OT among parents of children age 18–48 months, and association was stronger among males
Seltzer et al., 2014	CS	Humans	73	Physical abuse	Birth to present age (8 to 11 years)	CPS	In children, females with physical abuse history had higher urinary OT in response to a stressors compared to controls. Males with physical abuse history did not differ from controls.
Fries et al, 2005	CS	Humans	18	Extreme neglect	Infancy	O	Basal OT levels did not differ between children exposed to neglect and controls. After contact with their mother, OT levels increased for control children but not for children exposed to neglect. Sex differences not examined.
Vasopressin (AVP) (Lukas & Clutton-Brock, 2013; Tabak et al., 2015)
Lukas et al., 2010	E, P	Rats	95	Maternal separation	Up to day 14	N/A	Male rats exposed to maternal separation exhibit reduced AVP receptor expression in multiple various brain regions across development
Tabak et al., 2015	RCT	Humans	125	Lack of parental warmth	Up to age 16 years	RSR	AVP administration associated with greater empathic concern after viewing distressing and uplifting videos for individuals who reported more paternal warmth in childhood. No sexspecific effects.
Fries et al, 2005	CS	Humans	18	Extreme neglect	Infancy	O	Basal AVP levels were lower in children exposed to neglect compared to controls. Sex differences not examined. Sex differences not examined.
Glucocorticoids (Desantis et al., 2011; Kaess et al., 2018; Miller et al., 2017; Negriff et al., 2015; Pauli-Pott et al., 2017; Plotsky & Meaney, 1993; Wang et al., 2012)
Plotsky & Meaney, 1993	E, P	Rodents	30	Maternal separation	Postnatal days 2–14	N/A	Adult male rats exposed to long separations had increased plasma corticosterone responses to stress compared to male rats exposed to daily handling.
Wang et al., 2012	E, P	Mandarin voles	35	Paternal deprivation (removal of father at birth)	Up to postnatal day 21 (weaning)	N/A	Paternal deprivation associated with higher serum corticosterone levels in juvenile males
Negriff et al., 2015	L	Humans	454	Maltreatment	Birth to present age (9 to 13 years)	CPS	Childhood maltreatment history associated with lower salivary cortisol in response to an experimental stressor, but only among males
Miller, et al., 2017	P, L	Humans	99	Life stress (e.g. family death, financial problems)	Age 10 to 13 years	SR	Life stress in early adolescence was associated with attenuated CAR in late adolescence (16 to 19 years), but only among males.
Kaess, et al., 2018	L	Humans	69	Maltreatment	Up to age 14 to 16 years	RSR	Maltreatment history was associated with decreased CAR, but only among females.
Desantis et al., 2011	CS	Humans	39	Early life trauma (e.g. maltreatment, natural disaster)	Birth to age 18	RSR	In adults, ELA was positively associated with baseline serum corticotropin in females but negatively associated with baseline corticotropin in males. No sex differences in corticotropin response to an experimental stressor.
Pauli-Pott et al., 2017	CS	Humans	122	Family adversity (e.g. overcrowded living)	Up to present age (4 to 5 years)	PR	ADHD symptoms were associated with lower hair cortisol levels among males, but only when family adversity was high

Note: CAR, Cortisol Awakening Response; ADHD, Attention Deficit Hyperactivity Disorder

Study Design: E=Experimental, P=Prospective, L=Longitudinal, CS=Cross-sectional

ELA Measure: CPS=Child Protective Services records; PR=Parent report; RSR=Retrospective self-report; O=observation

4.2. Oxytocin and Vasopressin

The neuropeptide oxytocin (OT) plays a critical role in parenting as it promotes the nurturing, social sensitivity, and attunement necessary for child development (Carter, 2014). Another neuropeptide, vasopressin (AVP), is also associated with affiliative behavior, and may play a role in male social bonding and defensive or territorial behavior (Rajhans et al., 2019). Animal studies reveal that the OT and AVP systems are influenced by different early social experiences, including expression of OT and AVP receptors in the brain, and OT and AVP receptor expression is either increased or decreased depending on the type of ELA, the species studied, the sex of the offspring, and brain regions examined (Bales & Perkeybile, 2012; Veenema, 2012). Notably, male rats exposed to maternal deprivation exhibit reduced OT and AVP receptor expression in various brain regions, and both male and female rodents reared by single mothers or by mothers showing reduced care have reduced OT receptor expression in many brain regions (see Bales and Perkeybile, 2012). In a seminal study conducted with humans, Wismer Fries and colleagues found that children exposed to severe early deprivation had altered OT and AVP levels compared with family-reared controls (Fries et al., 2005). Other studies of human OT suggest that ELA is also associated with OT levels in adulthood, but the direction of this relationship is unclear. In a study of parents with young children, history of childhood physical abuse was associated with parents’ increased urinary OT, and this relationship was stronger among males than females (Mizuki & Fujiwara, 2015). However, in a study of adult males, retrospective stressful childhood experiences (e.g. abuse, neglect, loss of a relative) were associated with lower plasma OT (Opacka-Juffry & Mohiyeddini, 2012). It is possible that the direction of these relationships might be explained by differences between parents and non-parents, but family background was not reported for male participants in the latter study, and thus this comparison cannot be made.

Very little is known about the effects of ELA on AVP among humans, and we identified only one study examining AVP in adulthood as related to early childhood experiences. Tabak et al (2015) found that among individuals who reported more paternal warmth in childhood, AVP administration was associated with reporting greater empathic concern after viewing distressing and uplifting videos (see Table 4).

4.3. Glucocorticoids

Glucocorticoids, such as cortisol and corticosterone, play a complex role in the physiologic stress response and are commonly measured as indicators of hypothalamic-pituitary-adrenal (HPA) axis functioning (Nicolaides et al., 2015). Seminal studies conducted with rodents as early as the 1950s demonstrate that repeated brief separations of pups from their mothers result in decreased corticosterone responses to stress during adolescence and adulthood (Levine, 1957), while long separations in the first weeks of life are associated with increased corticosterone responses to stress in adulthood (Plotsky & Meaney, 1993).

In humans, an abundance of evidence indicates that there are sex differences in the relationship between ELA and HPA axis functioning, though recent systematic reviews report evidence for both male and female vulnerability (Carpenter et al., 2017; Gifford & Reynolds, 2017; Hollanders et al., 2017). For example, in a study of adolescents, Negriff et al (2015) found that childhood maltreatment history was associated with lower salivary cortisol among males, but not females, while Kaess and colleagues (2018) found past maltreatment history was associated with cortisol awakening response only among females. While disparate findings in adolescents may reflect physiological differences related to pubertal development, these findings are consistent with mixed evidence for sex differences in HPA-axis functioning in response to ELA noted in studies of children and adults (Desantis et al., 2011; Pauli-Pott et al., 2017).

5. Relationships between Males’ ELA and Gene-Environment Interactions and Epigenetics

5.1. Gene-environment (GxE) Interactions

Gene-environment interactions describe differential responses to an environmental exposure, such as ELA, based on an individual’s genotype (Nugent et al., 2011). As outlined in Table 5, recent evidence suggests that gene-environment interactions in response to childhood maltreatment exposure may differ between males and females (Maglione et al., 2018). For example, in a study of school-age children, Cicchetti and colleagues (2014) found that maltreatment history interacted with OXTR (oxytocin receptor gene) and FKBP5 (a gene involved in glucocorticoid receptor regulation) genotypes to predict child borderline personality symptoms, but the direction of the interaction differed by sex: females were at greater risk when they had minor (less common) alleles and males were at greater risk when they had major (more common) alleles. Other studies suggest that some genotypic vulnerability in response to ELA may be specific to males (Maglione et al., 2018). In a study of young adults, Min and colleagues (2017) found retinoid-related orphan receptor alpha (RORA) gene polymorphisms interacted with childhood maltreatment to predict anxiety sensitivity in boys. Other studies demonstrate monoamine oxidase A (MAOA; an enzyme involved in breaking down neurotransmitters such as dopamine and serotonin) genotype interacts with maternal parenting stress to predict internalizing behaviors in males at age 4 (Liu et al., 2017) and with history of childhood physical and emotional abuse to predict aggressive behavior among males in adolescence (Zhang et al., 2016).

Table 5.

Example Evidence of the Effects of Early Life Adversity on Males’ Gene x Environment Interactions & Epigenetics

Citation	Study Design	Species	N	Early Life Adversity (ELA)			Key Findings
				Type	Timing	Measure
Gene-Environment Interactions
Bennett et al., 2002	E, P	Rhesus macaques	132	Maternal separation	Up to 7 months postnatal	N/A	In males and females, adolescents with the short allele for rh5-HTTLPR who were exposed to peer-rearing (vs. mother-reared controls) showed significantly lower CSF concentrations of a serotonin metabolite (5-HIAA) and engaged in less play and more aggressive behaviors than peer-reared monkeys without the short allele
Newman et al., 2005	E, P	Rhesus monkeys	45	Maternal absence in infancy	Up to 7 months postnatal	N/A	Mother-reared adolescent males with the low-activity-associated MAOA genotype exhibited higher aggressive responses during competitive interactions and in typical social interactions
Cicchetti et al., 2014	CS	Humans	1051	Maltreatment	Birth to age 12	CPS	Maltreatment history interacted with and OXTR and FKBP5 genotypes to predict child borderline personality symptoms, but females were at greater risk when they had minor alleles and males were at greater risk when they had major alleles.
McQuaid et al., 2019	CS	Humans	475	Maltreatment	Childhood	RSR	Childhood maltreatment was associated with depressive symptoms for females regardless of IL-1β SNP genotype, while this relationship was strongest for males carrying the GG IL-1β SNP genotype
Min et al., 2017	CS	Humans	205	Maltreatment	Up to age 16 years	RSR	In young adults, RORA gene polymorphisms interacted with childhood maltreatment to predict anxiety sensitivity only in males
Zhang et al., 2016	CS	Humans	507	Physical and emotional abuse	Up to age 16 years	RSR	In adolescents, childhood physical and emotional abuse reported interacted with MAOA genotype to predict aggressive behavior among males
Epigenetics
Weaver et al., 2004	E, P	Rats	24	Cross-fostering to high- or low-maternal care	Up to day 21	N/A	Variations in maternal care resulted in changes in DNA methylation in both male and female offspring; effects persisted into adulthood and were reversed by cross-fostering
Checknita et al., 2020	CS	Humans	194	Physical or sexual abuse	Childhood	RSR	MAOA genotype interacted with child maltreatment history to predict aggressive behaviors in adult males, and the highest levels of aggression were found among males with high levels of MAOA methylation
Cicchetti et al., 2016	CS	Humans	548	Maltreatment	Up to current age (Mean 9.4 years)	CPS	Methylation of ALDH2 was higher in maltreated compared to non-maltreated males, but was lower in maltreated compared to non-maltreated females
Essex et al., 2013	L	Humans	109	Parental stress	Infancy and preschool	PR	High parental stress during early childhood was associated with differential DNA methylation at age 15, and patterning of epigenetic marks differed by child sex
Labonté et al., 2012	Postmortem	Humans	41	Abuse	Childhood	NR	Compared with non-abused controls, adult males with a history of abuse had differential methylation in promoter regions of several genes in hippocampal neurons
McGowan et al., 2009	Postmortem	Humans	36	Maltreatment	Childhood	Proxy-based interview	Hippocampal NR3C1 gene expression was decreased in maltreated individuals compared with non-maltreated controls
Roberts et al., 2018	CS	Humans	34	Maltreatment	Up to age 18	RSR	Childhood maltreatment was associated with sperm DNA methylation in 12 DNA regions, including sites on genes associated with neuronal function, fat cell regulation, and immune function

Note: SNP, Single nucleotide polymorphism

Study Design: E=Experimental, P=Prospective, L=Longitudinal, CS=Cross-sectional

ELA Measure: CPS=Child Protective Services records; NR=Not reported; PR=Parent report; RSR=Retrospective self-report

5.2. Epigenetics

Epigenetics refers to modifications to RNA or DNA that do not affect the DNA sequence, but can affect gene activity and therefore downstream outcomes of this activity (called endophenotypes; US National Library of Medicine, 2020). Thus, childhood experiences, including ELA, can affect gene function without altering the genes themselves. This epigenetic effect was demonstrated first in animal models and subsequently in studies with humans. In a 2004 study of rats, Weaver et al. were the first to demonstrate that variations in maternal care resulted in epigenetic differences, i.e., changes in DNA methylation, in both male and female offspring; they also demonstrated that these effects persist into adulthood. Subsequent research in animals has demonstrated that ELA results in differential DNA methylation patterns in the brain and peripherally (e.g., T cells in the blood; for review see Szyf & Bick, 2013).

Among humans, studies demonstrate DNA methylation is associated with ELA, including experiences of childhood abuse and family psychosocial adversity (e.g., financial stress, parental depression; Bush et al., 2018; Labonté et al., 2012; McGowan et al., 2009). Germ cells can also be affected by trauma exposure, and thus paternal transmission of stress, including history of childhood abuse, may occur through epigenetic changes in sperm (Roberts et al., 2018; Rowold et al., 2017; Yehuda & Lehrner, 2018). For further review on the epigenetic effects of ELA and transgenerational transmission of stress and adversity, see (Maccari et al., 2014; Scorza et al., 2019).

Epigenetic effects of ELA have been identified in a number of genes involved in stress-response pathways, including the HPA axis and oxytocin signaling (Burns et al., 2018; Hoffmann & Spengler, 2012; Kundakovic et al., 2013). Methylation may also affect the strength of gene-environment interaction effects. In a study of adults, MAOA genotype interacted with child maltreatment history to predict aggressive behaviors among males, and the most aggression was found among males with high levels of MAOA methylation (Checknita et al., 2020). Limited evidence also suggests sex-specific effects of ELA on DNA methylation (Table 5). For example, in a study of school-aged children, DNA methylation of ALDH2, a gene associated with risk for alcohol use disorders, was higher in maltreated males compared to non-maltreated males, but lower in maltreated females compared to non-maltreated females, suggesting a potential sex-specific pathway of epigenetic risk (Cicchetti et al., 2016).

6. Relationships between Males’ ELA and Behavior and Development

6.1. Externalizing Behaviors and Mental Health

Prospective animal studies that model certain aspects of mental health indicate that ELA may result in more aggressive, more anxious, and less social male offspring (Table 6; e.g., Maccari et al., 2014; Suomi, 1991, 1997). Animal studies also provide insight into biological mechanisms linking ELA and these behaviors, including the role of HPA axis activity and AVP and serotonin functioning (see Veenema, 2009, for a review). In humans, recent studies suggest males may be at potentially greater risk for exhibiting externalizing behaviors across development such as hostility and aggression (Gauthier-Duchesne et al., 2017; Logan-Greene & Semanchin Jones, 2015). For example, Terrell and colleagues (2018) found that early life stress (e.g., caregiver substance abuse, poverty, maltreatment) prior to age 3 was associated with violent behaviors in early adolescence among males, but not females. Studies conducted in adolescence and adulthood also suggest that males who experienced ELA are at greater risk than females for delinquency (Leban & Gibson, 2020), suicidal ideation (Seff & Stark, 2019), and intermittent explosive disorder (Puhalla et al., 2020). While some evidence suggests females are also at risk for externalizing behavior problems in response to ELA (Romano et al., 2016), sex differences may be attributable to the type of adversity experienced or the timing of behavioral assessment (Cyr et al., 2017; Godinet et al., 2014).

6.2. Attachment

Research in animals provided the foundational evidence that a secure attachment early in development has lifelong effects on cognitive and emotional health (Harlow & Harlow, 1965; Harlow & Zimmermann, 1959; see Hofer, 1996 for a review), and subsequent work has demonstrated that attachment is phylogenetically conserved across species (i.e., chicks, rodents, monkeys; Savidge & Bales, 2020; Sullivan et al., 2011). In a recent exploratory study of titi monkeys (Plecturocebus cupreus), a species in which infants form specific attachments to their fathers and adults form male-female pair-bonds (Spence-Aizenberg et al., 2016), monkeys that experienced ELA as infants (defined as experiencing the loss of a parent, a traumatic injury, or a significant separation from their attachment figure in infancy; Savidge & Bales, 2020) exhibited dissimilar infant attachment and adult attachment behavior patterns. This is in contrast to titi monkeys that did not experience ELA, for which infant attachment behaviors were correlated with adult attachment behaviors (Savidge & Bales, 2020). Of note, no sex differences were observed, though the small sample size (n=25) may have precluded these observations.

In human studies, ELA is associated with insecure attachment in childhood (Cyr et al., 2010), with studies indicating increased risk among male children. In a seminal study, Carlson and colleagues (1989) found not only significantly higher rates of disorganized attachment among maltreated children compared with non-maltreated children, but also that males were more likely than females to have a disorganized attachment pattern, regardless of maltreatment status (Carlson et al., 1989). In a study of mothers with infants, Lyons-Ruth and colleagues (1999) found that among infants of mothers with a history of maltreating or disruptive behaviors, disorganized attachment was more common among male infants than female infants (Lyons-Ruth et al., 1999). Other studies demonstrate that exposure to maternal psychopathology (Beeghly et al., 2017; Murray, 1992) and violence (Kobulsky et al, 2016), are associated with increased risk of insecure attachment in male children.

6.3. Substance Misuse

ELA, measured retrospectively, is also associated with problematic alcohol use and illicit drug use (Hughes et al., 2017), which may begin in adolescence or earlier (Löfving-Gupta et al., 2018; Ramos-Olazagasti et al., 2017). Research demonstrates increased risk for both males (Dragan & Hardt, 2016; El Mhamdi et al., 2017) and females (Meng & D’Arcy, 2016; Williams et al., 2020)); however, the pathways between ELA and substance misuse may differ by sex. In a study of adolescents, an association between childhood sexual abuse and problem drinking was mediated by anger symptoms among males, but mediated by anger and anxiety symptoms among females (Hudson et al., 2017). The association between childhood trauma, PTSD symptoms and problematic alcohol use may also be stronger for males (Rasche et al., 2016), and alcohol use may heighten aggression in males compared to females (Hoaken & Pihl, 2000).

Animal studies offer mechanistic evidence for the association between ELA and alcohol consumption. In a prospective study of young adult male monkeys that had been exposed to ELA or typical rearing in infancy, ELA monkeys consumed more alcohol in a free-drinking paradigm than controls, and also exhibited lower serotonin functioning in the brain (Huggins et al., 2012). Serotonergic functioning is associated with impulsivity, aggressive behavior, and substance use disorders (Sachs & Dodson, 2017).

7. Implications for Fathers’ Parenting

7.1. ELA and Fathers’ Parenting: Neurobiology and neurocognitive functioning

Neuroimaging studies demonstrate altered neural structure and function in individuals with histories of ELA (McCrory et al., 2017), and these alterations often occur in brain regions also implicated in parenting (see Table 1). Many complex cognitive (e.g., attention, executive functioning, and flexible thinking) and affective (e.g. reward, emotion processing and emotion regulation) processes that are critical for parenting undergo protracted development during childhood, and thus are also vulnerable to the effects of ELA (Lonstein et al., 2015; McCrory et al., 2017; Pechtel & Pizzagalli, 2011). Evidence suggests males may be particularly vulnerable to alterations in emotion processing, inhibitory control and limited emotion regulation strategies. Alterations in emotion processing, including reactivity to threatening stimuli or fearful facial expressions, coupled with diminished reward processing, may lead to the development of avoidance behaviors (Teicher et al., 2016) and could in turn lead to negative parenting among fathers. Impaired executive functioning and difficulties with emotion processing and regulation among males may also be particularly harmful for fathers’ parenting, as these skills are critical for providing sensitive caregiving, scaffolding to support child development, and coping with stressors related to parenting. Further, altered neural functioning in emotion regulation circuitry may increase risk for paternal psychopathology and disrupt a father’s ability to manage the stresses of parenting. This increased risk among males is supported by a recent study of parental burn-out, in which burned out fathers were more likely than mothers to experience escape and suicidal ideations and to exhibit neglectful behaviors toward children (Roskam & Mikolajczak, 2020).

Similar neural networks are activated by mothers and fathers in response to their children, but developmental timing and recruitment of these networks differs by sex (for review, see Rajhans et al 2018). In contrast to females, who experience prenatal neurobiological changes in response to pregnancy (Abrahams and Feldman, 2018), neuroplasticity among males is primarily driven by postnatal social experiences with parenting (Kim et al., 2014), suggesting this may be an important window of opportunity for intervention among fathers who experienced ELA. However, neuroplasticity in fathers is not well understood. Longitudinal functional neuroimaging studies in humans, along with basic science studies in animal models, are required to understand neural structural changes associated with ELA may impact neuroplasticity during the transition to fatherhood. It is also important to note that neural structure conditions, but does not determine, neural functioning (Batista-García-Ramó & Fernández-Verdecia, 2018).

7.2. ELA and Fathers’ Parenting: Hormones and Hormone Receptors

Fatherhood induces hormonal changes in males, including hormones involved in the hypothalamic-pituitary-adrenal (HPA) axis, which serves as the body’s primary stress-response system, and hypothalamic-pituitary-gonadal (HPG) axis, which is involved in reproductive and immune system regulation (Table 2). For example, the transition to fatherhood is associated with decreases in testosterone and glucocorticoids, increases in prolactin, and variations in estradiol, oxytocin, and vasopressin (Bales & Jarcho, 2013; Bales & Saltzman, 2016; Rilling & Mascaro, 2017; Saltzman & Ziegler, 2014; Storey & Walsh, 2011). While limited, evidence from human and animal studies suggest that these same hormones are also affected by ELA, but the direction of effects may differ: ELA is associated with increased testosterone levels, decreased oxytocin levels, and cortisol dysregulation. It is possible that effects on testosterone and cortisol increase risk for aggressive parenting behaviors among males who experienced ELA (Montoya et al., 2012), or that effects on oxytocin may affect fathers’ mental health or father-child interactions, as oxytocin plays a role in regulation of stress and anxiety (Kormos & Gaszner, 2013). However, as very few studies have examined relationships between ELA and these hormones among human fathers, it is currently unclear whether and how hormonal changes during the transition to parenthood are affected by fathers’ ELA history. Further, while most studies examined hormones in isolation, interactions between these hormones likely play an important role in influencing fathering behavior, and research is required to better understand these relationships (Rajhans et al., 2019). Given the enormous downstream consequences of OT and AVP receptor density on social functioning (Bale et al., 2001; Ross et al., 2009), studies of hormone receptors using animal models also represent an import avenue for future research on mechanisms underlying the link between ELA and later parenting in males.

There is also a surprising dearth of studies examining glucocorticoids among fathers who experienced ELA, despite evidence for a role of cortisol in human parenting behavior (Bos, 2017) and for sex differences in HPA-axis functioning in response to ELA (Carpenter et al., 2017; Gifford & Reynolds, 2017; Hollanders et al., 2017). Limited evidence suggests early life programming of the HPA axis may affect the quality of fathers’ interactions with their children or abilities to cope with stressful situations related to parenting; however, it is unclear whether males or females are more vulnerable to effects on HPA axis functioning following ELA. Empirical human and animal studies conducted with fathers, and comparing fathers to mothers, are necessary to better understand the influence of HPA-axis dysfunction on parenting among fathers who experienced ELA.

7.3. ELA and Fathers’ Parenting: Gene-environment (GxE) Interactions and Epigenetics

While studies on genetics of human fathering are limited (Mileva-Seitz et al., 2016; Neiderhiser et al., 2007), evidence for sex differences in gene-environment interactions and epigenetic responses to ELA suggests that ELA may have important effects on fathers’ parenting, and that these may differ from effects on mothers’ parenting. Further, animal studies demonstrate that ELA may not only affect offspring through influences on fathers’ parenting behaviors, but also through epigenetic programming, a key component of the concept described as the “Developmental Origins of Health and Disease” (Goyal et al., 2019; Gluckman & Hanson, 2006; Soubry, 2018). For example, in rodents, fathers can transmit environmentally-induced effects to offspring, including offspring phenotypes that are influenced by the father’s exposure to stress, nutrition, and toxins (Braun & Champagne, 2014). These inherited phenotypes are thought to be transmitted epigenetically via the patriline, meaning that epigenetic information may be inherited from fathers (D’Urso & Brickner, 2014), and may be an important mechanism underlying the intergenerational transmission of ELA. Although the influence of epigenetic programming on human fathers’ parenting remains unclear, evidence from rodents suggests that epigenetic mechanisms may contribute to transmission and variability in parenting (Jensen Peña & Champagne, 2012). Emerging and innovative areas of research, including the epigenetic transmission of adversity through sperm (Roberts et al., 2018), and the role of plasticity alleles in transmission of parenting behavior (Beaver & Belsky, 2012), may also help elucidate genetic and epigenetic influences on parenting among fathers who experienced ELA.

Candidate gene studies suggest that interactions between genotype and ELA may differ by sex, and that males may be particularly vulnerable to poor outcomes. Though replication of these studies is required, preliminary evidence suggests that male-specific gene-environment interactions, such as the link between MAOA and aggressive behaviors, may provide valuable insight into genotypic risk factors that can be targeted using precision health approaches. Future examination of candidate genes implicated in mothering, including dopamine, vasopressin, and serotonin genes, as well as the development of polygenic risk scores, may provide insight into GxE interactions that influence parenting among fathers who experienced ELA (Wertz et al., 2019).

7.4. ELA and Fathers’ Parenting: Behavior and Development

The clearest evidence for potential effects of ELA on fathers’ parenting is derived from studies of males’ behavior and development. Evidence suggests that males may be more vulnerable to attachment disorders and externalizing behaviors in response to ELA, both of which have important implications for later behaviors that may influence parenting, including aggression, substance misuse, and mental illness. Of significant concern is the increased risk for disorganized attachment among males exposed to ELA, as this is not only associated with externalizing behaviors, but also a risk for later psychopathology and child maltreatment behaviors (Fearon et al., 2010; Fulu et al., 2017; Nakash-Eisikovits et al., 2002; Schimmenti et al., 2014), Also of concern is the potential for externalizing and aggressive behaviors in childhood to lead to later violence, including intimate partner violence (IPV) and child maltreatment, thus perpetuating cycles of family trauma and adversity. In a multi-country study conducted by the United Nations, all forms of childhood trauma (e.g. maltreatment, witnessing IPV) were associated with IPV perpetration (Fulu et al., 2017), and other studies have found this relationship may be mediated by emotion dysregulation and substance misuse, outcomes also associated with ELA exposure among males (Brown et al., 2015; Gratz et al., 2009).

Despite these behavioral risks among males exposed to ELA, there is insufficient evidence to draw conclusions regarding implications for parenting among fathers who experienced ELA. This is because the transition to fatherhood is associated with neurologic, hormonal, and epigenetic changes, which may also influence behavior and development, including behaviors associated with ELA exposure. There are also a myriad of other environmental factors that may influence fathers’ behavior; for example, risk for IPV is also associated with current poverty-related stress (Jewkes, 2002). Further, emerging evidence suggests fathers’ involvement with their children may be protective against poor outcomes (Brown, 2021). Thus, in order to empirically understand how the behavioral outcomes associated with ELA among males (e.g. insecure attachment, externalizing behaviors, substance misuse) actually influence fathers’ parenting, research must be conducted with fathers directly.

8. Discussion

Fathers have a unique and important influence on children’s development, and yet factors that influence fathers’ parenting have been vastly understudied. In order to advance understanding of the relationship between ELA and fathers’ parenting, we integrated knowledge from animal models and human studies across a range of scientific disciplines to examine the effects of ELA specifically among males. The results of our review suggest that like mothers, history of ELA among fathers may influence parenting through both biological and behavioral pathways. Our conceptual model (Figure 1) is intended as an extension of Cabrera et al.’s (2014) expanded model of fathering to specifically examine the effects of fathers’ ELA history. The physiological and behavioral pathways presented in this model and the gaps identified in this review represent potential avenues for future research in both animal and human samples. We expect this model will be refined and expanded upon as additional empirical studies become available.

While more empirical studies with fathers are needed, findings of this review suggest that interventions developed to support mothers who experienced ELA may be insufficient or ineffective for fathers who experienced ELA. For example, increased risk for disorganized attachment, aggressive behaviors, and poor inhibitory control among males may require interventions focused on emotion regulation and mentalization (Stover et al., 2020). This approach differs from parenting interventions designed for mothers, which often focus on attachment and synchrony (Davis et al., 2018). Until additional empirical studies on ELA and fathers’ parenting become available, our conceptual model and the findings from this review can be used to identify risk factors and potential targets for intervention among fathers who experienced ELA.

In our conceptual model, we present male neurobiology, hormones, and genetics as separate biological pathways that influence males’ behavior and development, and ultimately influence fathers’ parenting. However, these biological pathways also interact with one another, and these interactions may influence behaviors in ways that are not yet well understood (Cabrera et al., 2014). It is also unclear whether the experience of fatherhood, which includes neuroplastic and hormonal changes, may alter the effects of ELA on biology or behavior. For example, it is unknown whether the increased testosterone associated with ELA prevents a decrease in testosterone during the transition to fatherhood, and thus leads to more aggressive behaviors, or whether the decreased testosterone associated with fathering protects against the effects of ELA on testosterone levels. Additional research with fathers, especially during the transition to parenthood, may provide important insight into these relationships.

While our review was largely focused on understanding the harmful effects of ELA among males, there is also a need to study resilience and protective factors among fathers who experienced ELA. Past studies with mothers demonstrate that protective factors during childhood, including supportive caregivers and positive peer relationships, may buffer the harmful effects of ELA, and as such, mothers with a history of family strengths engage in more supportive, reflective parenting (Condon et al., 2021; Narayan et al., 2020). It is also important to note that recent research suggests ELA predicts poor outcomes on a group level, but not an individual level (Baldwin et al., 2021). We highlight childhood protective factors in our conceptual framework to acknowledge the importance of these experiences, especially in explaining individual differences in outcomes. Examination of childhood protective factors specifically among fathers, as well as examination of sex differences in protective mechanisms, may provide novel insight necessary to inform the development and targeting of parenting interventions.

Characteristics of ELA, including the type, timing, and severity of experiences, are known to have differential effects on later outcomes (Oh et al., 2018). Though evidence is limited, findings of the current review suggest that these characteristics may influence the pathways outlined in our model and therefore potentially influence later parenting among fathers. For example, ELA in early childhood, but not adolescence, was associated with decreased OT levels in adulthood (Opacka-Juffry & Mohiyeddini, 2012). Further, the type of adversity experienced may help explain sex differences in responses to ELA. For example, Cyr and colleagues (2017) found that sexual victimization history predicted anger and aggressive behavior among adolescent females, but physical assault history predicted anger among adolescent males (Cyr et al., 2017). Studies of neurocognitive functioning also indicate that the effects of ELA may differ based on sex and adversity type, as stress perception was increased among females with a history of threat-based adversity (e.g., abuse), but among males with a history of deprivation (e.g., neglect; LoPilato et al., 2020; McLaughlin et al., 2014). Empirical studies are needed to understand how the type, timing, and severity of ELA experiences influence parenting, and whether and how these differences vary based on sex.

While research on fathers is growing in the scientific literature, there remains a need to better understand both influences on and effects of parenting by human fathers (Cabrera et al., 2014; Cabrera et al., 2018). Just as with mothers, there is also need to conceptualize the father role beyond biological fathers and consider caregiving by other male figures such as uncles, grandparents, step-parents, and adoptive parents (Cabrera et al., 2018). Future research with males in diverse father roles will improve understanding of fathering behavior and outcomes for children. Future research studies must also consider the complexity of family relationships and the influence this may have on fathers’ parenting. For example, fathers are known to effect children both directly through parenting behaviors, and indirectly through effects on maternal stress, maternal mental health, and both instrumental and emotional maternal support (Condon & Sadler, 2019; Sarkadi et al., 2008). Limited evidence also suggests fathers may protect children against the harmful effects of stressors such as maternal depression (Cabrera et al., 2014; Lewin et al., 2014). Furthermore, triadic relationships between the mother, father, and child may also have important psychological effects that influence neurobiology and hormone functioning among males, which may in turn effect both offspring development and fathers’ parenting (De Mendonca et al., 2019; Gordon & Feldman, 2008). Therefore, considering triadic relationships between the father, mother, and child is an important direction for future research.

Research with both humans and non-human animals is needed to address gaps in the literature and should be used as appropriate to answer specific research questions. For example, regarding research into triadic relationships, certain animals will be stronger models than others: in nonhuman primates, for instance, triadic relationships as defined above cannot be studied in macaques (Macaca spp., which exhibit maternal care and no pair-bonding) but may be well-suited for other primates such as titi monkeys and marmosets (Callicebus and Callithrix spp., which exhibit biparental care and do form pair-bonds). Studies with non-human animals can also be used to further our understanding of the influence of ELA on hormones among males, particularly testosterone and oxytocin. Experimental studies with non-human animals can also be designed to determine the biological mechanisms of fathers’ parenting, such as patterns of gene expression and interactions with early rearing environments, to test hypotheses regarding the regulation of fathers’ parenting behavior. In humans, future longitudinal studies could provide insight into the role hormones play in the structural and functional neural changes seen over the transition to fatherhood among fathers who experienced ELA (Kim et al., 2014). Human studies may also be used to prospectively examine how the biological adaptations associated with ELA (e.g., hormone levels) influence later parenting, and whether and how sex differences in response to ELA (e.g., neural structure and gene-environment interactions), translate to sex differences in parenting behaviors. Such work could identify targets for potential interventions.

Strengths & Limitations

A major strength of this review is our comprehensive and multidisciplinary approach to understanding the effects of ELA among fathers, a vastly under-studied but highly important group. Another strength is our inclusion of animal literature to describe foundational scientific discoveries or to understand biological mechanisms that cannot be easily studied in humans. However, translation of nonhuman animal studies is limited, and in nonhuman animal species just as in humans, we must consider the life histories, species differences, and evolutionary history that results in both the conservation of traits (such as fathering behavior) as well as variations between individuals and across groups in behavioral manifestations of these traits (such as differences in the degree and types of paternal care behaviors; Gettler, 2014).

This review was also subject to a number of other limitations. Given the infancy of this field, we defined ELA broadly in order to capture the breadth of postnatal influences on males’ biology and behavior. However, while we described the ELA characteristics examined whenever possible, given the wide variance in ELA definitions and measures, we were unable to draw conclusions about the influence of ELA characteristics such as type, timing, and severity. It was also beyond the scope of this review to discuss biological mechanisms and pathways in detail, as well as to examine the influence of prenatal ELA exposure, which is also known to influence offspring biology and development (Beijers et al., 2014; Graignic-Philippe et al., 2014). Finally, we did not explore the role of childhood protective factors or current environmental stressors, such as the family environment or relationships with other caregivers, which may influence developmental trajectories and parenting behaviors. However, we highlight these in our conceptual model to emphasize the importance of considering these factors in future studies.

8. Conclusion

Fathers have an important and unique influence on child development, and father involvement is associated with positive social, behavioral, and psychological outcomes for both mothers and children (Sarkadi et al., 2008; Yogman et al., 2016). The conceptual model and evidence presented in this review can be used as a framework to direct future biobehavioral research on parenting among males who experienced ELA, improve understanding of intergenerational transmission of adversity, and develop targeted interventions with a paternal or biparental focus.

Highlights.

• Early life adversity (ELA) may influence fathers’ parenting, but empirical evidence is limited.

• Males exposed to ELA are at increased risk for poor emotion regulation, aggressive behaviors, and insecure attachment patterns.

• Research is needed to understand effects of ELA on males’ hormones during the transition to fatherhood.

• Our conceptual framework may direct future research and interventions for fathers with ELA history.