Prenatal Stress and the Developing Brain: Postnatal Environments Promoting Resilience

Heightened maternal stress during pregnancy is associated with atypical brain development and an elevated risk for psychopathology in offspring. Supportive environments during early postnatal life may promote brain development and reverse atypical developmental trajectories induced by prenatal stress. We reviewed studies focused on the role of key early environmental factors in moderating associations between prenatal stress exposure and infant brain and neurocognitive outcomes. Speci ﬁ cally, we focused on the associations between parental caregiving quality, environmental enrichment, social support, and socioeconomic status with infant brain and neurocognitive outcomes. We examined the evidence that these factors may moderate the effects of prenatal stress on the developing brain. Complementing ﬁ ndings from translational models, human research suggests that high-quality early postnatal environments are associated with indices of infant neurodevelopment that have also been associated with prenatal stress, such as hippocampal volume and frontolimbic connectivity. Human studies also suggest that maternal sensitivity and higher socioeconomic status may attenuate the effects of prenatal stress on established neurocognitive and neuroendocrine mediators of risk for psychopathology, such as hypothalamic-pituitary-adrenal axis functioning. Biological pathways that may underlie the effects of positive early environments on the infant brain, including the epigenome, oxytocin, and in ﬂ ammation, are also discussed. Future research in humans should examine resilience-promoting processes in relation to infant brain development using large sample sizes and longitudinal designs. The ﬁ ndings from this review could be incorporated into clinical models of risk and resilience during the perinatal period and used to design more effective early programs that reduce risk for psychopathology.

Prenatal adversity, including elevated maternal stress and depressive and anxiety symptoms during pregnancy, has been consistently associated with an increased risk for psychopathology in children (1). Given that maternal stress during pregnancy is prevalent (2,3), designing ways of reducing this risk for women and children is critical. Maternal stress during pregnancy is believed to alter gestational stress biology, including neuroendocrine and inflammatory processes, and in turn affect fetal brain development, contributing to risk for psychopathology (4,5). Prenatal adversity has been linked to fetal and infant brain structure and function, such as smaller overall brain volume (6), altered cortical thinning (7), and altered corticolimbic structure, microstructure, and functional connectivity (8)(9)(10)(11)(12)(13). Given that a substantial amount of brain development takes place in utero, the prenatal period is a time of particular susceptibility to environmental influences (9,14).
Although prenatal stress increases risk for psychopathology in offspring, not all children with a history of exposure to prenatal adversity develop psychiatric disorders. Certain factors have been identified that modify the association between prenatal adversity exposure and infant brain development, contributing to variability in mental health outcomes. The early postnatal period, similar to the prenatal period, is characterized by enhanced neural plasticity (15,16) due to the rapid development of structural and functional neural networks (14). In humans, compelling evidence for early postnatal plasticity comes from studies of children exposed to early deprivation in orphanages who then transition to supportive family environments (e.g., through adoption). Although the negative developmental effects of early deprivation can persist over time, children often show remarkable developmental catch-up following placement in family environments, especially when they are placed at young ages (17)(18)(19). Supportive early postnatal environments may play a key role in moderating the associations between prenatal stress and infant brain development, reducing the risk for psychopathology in children.
After prenatal stress exposure, postnatal experience may affect the infant brain through various pathways. Elevated maternal stress during the perinatal period may interfere with the transition to parenthood (20) and the quality of caregiving provided to the infant (21) and consequently lead to an accumulation of risk exposures for the offspring. However, postnatal environmental factors may also moderate the associations between prenatal stress and brain structure and function in offspring (22). In nonhuman animal studies, factors such as sensitive caregiving and environmental enrichment have been found to reverse the effects of prenatal stress on brain development (22,23) and promote resilience (24). Early postnatal environmental factors may, for example, directly promote resilience and compensate for prenatal adversity, reverse negative effects that are prenatal in origin, or protect from altered developmental trajectories initiated by the prenatal environment (25).
In this narrative review, we examined whether early postnatal environmental factors (i.e., parental caregiving quality, environmental enrichment, social support, and socioeconomic status [SES]) (see Figure 1) measured within the first few years of postnatal life may promote resilient infant brain development following exposure to prenatal adversity. The focus is on studies examining postnatal environmental factors and infant brain outcomes after maternal stress during pregnancy. Where only a few studies of moderation or resilience processes directly measuring prenatal stress were available, we considered studies of measures that may reflect prenatal stress (e.g., low birth weight, preterm birth, elevated maternal cortisol levels) (26). Outcomes that are the primary focus of this review include not only infant brain structure and function but also infant neurocognitive performance and hypothalamic-pituitaryadrenal (HPA) axis functioning, which reflect mechanisms at the neural level.
Findings suggesting both protective and promotive effects are presented (24). Protective factors have variable effects depending on the level of risk, with greater influence at higher levels of risk or adversity, and are typically indicated by statistical interactions. Promotive factors refer to predictors of more adaptive outcomes regardless of risk level, similar to the main effect in statistical models. We include promotive effects that are demonstrated when a putative positive factor leads to favorable outcomes in a sample exposed to prenatal adversity (e.g., preterm birth, maternal depression during pregnancy).
Finally, we review the biological mechanisms that may underlie these effects, including oxytocin, epigenetic alterations, and inflammation. While genetic protective factors are important (24,27), they are beyond the scope of the present review.

PRENATAL STRESS AND THE DEVELOPING INFANT BRAIN
Animal studies have shown that prenatal stress exposure alters offspring brain structure and function, especially in the hippocampus and amygdala (28,29). These subcortical structures may be particularly susceptible to intrauterine glucocorticoid-mediated influences due to their high number of glucocorticoid receptors (29,30). Indeed, animal studies have linked prenatal stress exposure with reduced dendritic length and neurogenesis in the hippocampus (23,31,32). In humans, higher maternal stress during pregnancy has been associated with smaller newborn hippocampal volume (33) and slower growth of the hippocampus across the first 6 months of life (34).
Human infant magnetic resonance imaging studies have also reported associations between prenatal stress and frontolimbic structural and functional connectivity. In diffusionweighted imaging studies, prenatal stress has been associated with reduced structural connectivity in multiple fiber bundles in frontolimbic circuits in neonates (12,13,35,36). Increased maternal stress during pregnancy has also been associated with weaker infant frontolimbic functional connectivity (37)(38)(39). Importantly, these frontolimbic circuits play key roles in the regulation of the HPA axis (40), emotion processing and regulation (41), learning, and memory (42). Altered frontolimbic structure and function in infancy have been associated with an increased risk for later emotional and behavioral problems (10,43).

EARLY POSTNATAL ENVIRONMENTAL FACTORS
According to bioecological models, children's development is embedded in multiple layers of environmental influences, including those that are proximal and those that are distal to children's immediate surroundings (44). During infancy, the most salient proximal factors are often in the home environment. Proximal environmental factors, including caregiving quality and cognitive and linguistic stimulation in the home (44), are shaped by distal environmental factors, notably SES and social support provided to the family. Middle to high SES has been associated with early home environments characterized by more sensitive and responsive parenting and greater cognitive and linguistic stimulation compared with low SES (45,46).
Resilience-promoting postnatal factors that are the focus of this review (parental sensitivity, parental stroking or skin-toskin contact, cognitive stimulation/environmental enrichment, social support, SES) have been found to offset the negative effects of prenatal adversity exposure on children's internalizing, externalizing, and attention symptoms (47)(48)(49)(50)(51)(52)(53)(54)(55). For example, one study reported an association between maternal prenatal anxiety and child externalizing symptoms only when maternal sensitivity was low (54). Complementing correlational and longitudinal studies, intervention studies show that postnatal environmental factors (e.g., parental sensitivity) have positive associations with social-emotional and behavioral functioning in infants with indicators of prenatal stress exposure (56,57). These studies support the notion of causal associations, especially when coupled with evidence from animal models. Here, we discuss studies describing how resiliencepromoting processes may operate at the neural level through effects on the infant brain.

Proximal Environments
Parental Caregiving Quality. Consistent with nonhuman animal research (58,59), human studies have established associations between exposure to maltreatment (abuse, neglect) and brain structure and function in children (17,60,61). Animal models have also indicated that quality of maternal care (e.g., licking and grooming) is associated with greater dendritic length and spine density in the hippocampus and alterations in the synaptic plasticity and glucocorticoid receptor transcription in the hippocampus and amygdala (59,62,63). In human studies, higher-quality caregiving has been linked with volumes of infant limbic and prefrontal structures, connectivity of frontolimbic pathways, and activation patterns in socially relevant areas of the infant brain (64-68) and with higher frontal resting alpha and theta electroencephalography (EEG) power and frontal EEG asymmetry (64,69). Furthermore, caregivingfocused randomized controlled trials (RCTs) in infancy have shown long-term positive influences on brain structure and activity (61,70) and HPA axis regulation (e.g., decreased cortisol reactivity and faster recovery) (71-73).
Seminal work in rodents first demonstrated that high-quality maternal care (e.g., licking and grooming) might reverse the effects of deprived care environments on glucocorticoid receptor expression in the hippocampus and the paraventricular nucleus of the hypothalamus (74,75) and on neurocognitive performance (75,76). Similarly, maternal care can counteract prenatal stress-induced gene expression related to neuroendocrine signaling relevant for anxiety (22). Early postnatal handling (removing the pup from its mother for brief periods during early life), which leads to enhanced maternal care, can reverse the effects of prenatal adversity on neural proliferation, cell survival and differentiation in the offspring hippocampus, and gene expression relevant for shaping of presynaptic sites and neurotransmitter release (23,77).
Human studies have shown similar patterns of results. For example, taking a retrospective measure of human childhood bonding with a mother, one study found that lower birth weight was associated with smaller hippocampal volume only in adult women who reported poorer parental care, but not among those reporting higher-quality care (78). Beyond direct brain measures, several prospective studies have demonstrated postnatal moderation for infant HPA axis regulation and neurocognitive development. For instance, the association between maternal psychiatric illness during pregnancy and infant higher cortisol at baseline was not observed among infants receiving sensitive caregiving postnatally (79). In addition, maternal cortisol level during pregnancy has been associated with higher child cortisol reactivity only when maternal sensitivity was low, but not when maternal sensitivity was high (80). Infants exposed to high maternal anxiety and distress during pregnancy did not show poorer cognitive performance when maternal sensitivity was high (81,82). In addition, the association between low birth weight and poorer executive functioning at preschool age was offset when parental sensitivity in toddlerhood was high (83).
There is limited intervention evidence from populations exposed to prenatal stress, with the majority of prenatal and postnatal RCTs focusing on parental well-being or parenting behaviors as outcomes. However, RCT studies on preterm infants indicate that interventions targeting parental sensitivity or skin-to-skin contact (e.g., Kangaroo Care) are associated with more mature neonatal brain structural connectivity (84), left frontal EEG asymmetry (85), increased frontal EEG power in high-frequency bands (85,86), and decreased cortisol reactivity (85,87). Similarly, following such interventions, infants have shown decreased autonomic nervous system reactivity and improved cognitive and language development (57,88).

Cognitive and Linguistic Stimulation
In animal models, environmental enrichment or complexity (e.g., novel objects and social partners) has been found to lead to increased cortical thickness (89) and to promote dendritic branching, synaptic density, and higher rates of neurogenesis in the hippocampus (90,91) and in turn improve learning and memory (92). In humans, it is well established that early cognitive and linguistic stimulation (e.g., parental language input, toys and books in the home, reading to the child) supports infant cognitive and social-emotional development (93)(94)(95) significantly associated with brain structure and function in infants and young children, including in neural circuits responsible for language, attention, and self-regulation (96,97). In addition, greater language input has been associated with lower relative power in low-frequency bands and higher relative power in high-frequency bands in infant EEG studies (98). Translational models have shown that postnatal environmental enrichment can reverse the effects of prenatal stress on outcomes, including spine density of hippocampal neurons, synaptic plasticity in the cortex and hippocampus, and learning and memory in offspring (32,99). Consistent with these findings, human research suggests that high cognitive/linguistic stimulation may facilitate neurocognitive development in infants exposed to prenatal adversity. For example, research has indicated significant interactions between birth weight and responsive parenting (100,101). In one study, children with lower birth weights did not show decreases in their language skills when maternal responsiveness was high (100). Furthermore, an intervention that promoted maternal reading to infants in the neonatal intensive care unit had a positive impact on early reciprocal social interactions in infants born preterm (102), suggesting that infant-enhanced neurocognitive skills, including language, could compensate for the negative impacts of prenatal stress on social cognitive neural networks.
Taken together, the evidence indicates that parental sensitivity may promote adaptive HPA axis regulation, neurocognitive development, and structure and function in corticolimbic regions of the brain after exposure to prenatal adversity. Furthermore, cognitive stimulation has been associated with greater cortical and hippocampal gray matter and differences in connectivity in language and executive function networks. However, some caregiving-focused studies, including RCTs, have failed to detect patterns of results suggesting such resilience processes (103)(104)(105). Future research should embed measures of infant brain structure or function into prospective study designs and seek to disentangle the role of different postnatal caregiving factors in relation to prenatal stress exposures.

Parental Social Support
Positive social interactions and relationships are linked to more adaptive brain functioning across the lifespan (106). During the prenatal period, social support may influence maternal physiological stress reactivity (107), shaping the intrauterine milieu and in turn contributing to fetal neurodevelopment (108). Postnatally, social support received by the parent (e.g., perceived social support, marital relationship status and satisfaction) may promote their mental health (106), increasing parental warmth and sensitive caregiving and in turn facilitating infant brain development (109).
Social support has been identified as a moderator of associations between prenatal stress and infant brain development (110,111). For example, with low maternal social support, higher maternal depressive or anxiety symptoms during pregnancy were associated with weaker infant amygdalaprefrontal functional connectivity. In the context of high social support, no such associations were detected (111). Thus, atypical functional connectivity of emotion regulation circuitry in infants exposed to high prenatal stress may not be observed in infants whose mothers received high social support. Additionally, one study found no association between maternal prenatal diurnal cortisol output and infant cortisol reactivity at 5 to 10 months of age when the mother experienced high social support postnatally (108).
Social support may have effects on mothers and their infants through both prenatal and postnatal mechanisms. To date, few studies have focused specifically on the moderating effects of parental postnatal social support with regard to infant brain structure or function, but effects have been suggested for emotion and stress regulation circuitry in the infant brain.

Socioeconomic Status
Higher SES (e.g., family income, parental education) is a significant predictor of more adaptive cognitive development and better mental health in children (112,113). In neuroscientific studies, socioeconomic factors have been associated with indices of brain structure and function, both prenatally and postnatally, that underlie cognitive development and mental health (113). In children and adolescents, higher SES has been repeatedly linked with larger hippocampal volume (114). SES during the prenatal period has been associated with fetal (115) and neonatal (110,116) brain structure. During infancy, SES has been associated with differences in brain structure and development (117,118) and functional connectivity (119,120). For example, higher SES has been associated with greater cortical (e.g., in the frontal and parietal lobes) and subcortical gray matter volume in infants (117,118).
Furthermore, in EEG studies, higher SES has been linked with greater mid-to high-frequency (i.e., alpha, beta, gamma) band power and lower low-frequency band power in infants (121)(122)(123)(124). Moreover, an RCT showed that poverty reduction (unconditional cash transfer) in early life was significantly associated with greater EEG power in high-frequency bands in infants (125), supporting the notion that socioeconomic context has causal effects on infant brain function.
Few studies have examined the interplay between prenatal adversity exposure and SES in predicting infant brain structure or function. In one study, associations between higher SES at birth and increased resting-state functional connectivity within subregions of the ventrolateral prefrontal cortex and decreased functional connectivity between the striatum and frontopolar prefrontal cortex were present in both preterm and full-term newborns (120), consistent with the possibility of SES effects on functional connectivity in neonates born preterm. In contrast to the few neuroscientific studies, a considerable number of large-scale studies have yielded evidence suggesting that higher SES supports more optimal neurocognitive development in children born preterm or with low birth weight (126)(127)(128). Some studies have found statistical interactions (e.g., protective effects) suggesting that higher SES may reduce the negative effects of preterm birth on neurocognitive outcomes, including language and executive function, in early childhood (128)(129)(130)(131)(132). Other studies have also yielded findings suggesting that young children born preterm have more optimal neurocognitive outcomes in higher SES contexts (i.e., promotive effects; there were significant main effects of SES  (126,129,(133)(134)(135). Similar to the effects of social support, various proximal factors could explain the positive effects of higher SES on neurocognitive and brain outcomes after exposure to prenatal adversity. One possibility is that higher SES leads to a more optimal early postnatal home environment characterized by higher caregiving quality and more cognitive and linguistic stimulation, effects that have been described earlier in this review. In addition, protection conferred by higher SES may occur prenatally (114). For instance, greater maternal nutritional sufficiency or better maternal diet and exercise during pregnancy, often afforded by greater socioeconomic resources, may buffer against the effect of elevated maternal stress during pregnancy on fetal and infant brain development.

POTENTIAL BIOLOGICAL PATHWAYS UNDERLYING PROTECTIVE AND PROMOTIVE INFLUENCES
Mechanisms that may underlie these protective and promotive influences on the infant brain following exposure to prenatal stress include oxytocin, moderation of the epigenome relevant for the HPA axis, and inflammation. Oxytocin, a neuromodulator released during HPA axis activation (136), is a wellestablished biological factor underlying the formation of social bonds (106), including the parent-child relationship (137). Nonhuman animal studies suggest that maternal presence blocks HPA axis activation and increases oxytocin levels in infants (138). Both prenatal stress and caregiver sensitivity can affect the expression of offspring genes related to oxytocin pathways (139,140). In turn, oxytocin has the potential to prevent the negative effects of glucocorticoids and promote neuronal development in the hippocampus and inhibit anxiety and fear reactions by modulating amygdala and prefrontal cortex activity (141,142). Oxytocin also affects corticotropinreleasing factor gene transcription (143), an epigenetic pathway related to HPA axis reactivity. Therefore, oxytocinrelated pathways are relevant candidate mechanisms for caregiver-mediated protective effects on the infant brain.
For moderation of the epigenome relevant for the HPA axis, prior studies indicate that caregiver sensitivity and touch (physical affection) modulate the methylation of genes encoding glucocorticoid receptors (e.g., NR3C1), serotonin functions (e.g., SCL6A4), and brain-derived neurotrophic factor (i.e., BDNF) (139,144). In particular, increased methylation of NR3C1 has been associated with the extent of prenatal stress exposure (145), whereas less methylation has been related to higher-quality caregiving (146,147). Recent studies indicate that infants exposed to both high maternal depression symptoms and low sensitivity postnatally had higher methylation of NR3C1 than infants exposed to depressive symptoms but high maternal sensitivity (148). Children exposed to higher maternal postnatal depression showed higher rates of NR3C1 methylation, but this effect was reversed by frequent maternal stroking (149). NR3C1 methylation changes, mainly increased methylation, have also been linked to neurobehavioral and neuroendocrine outcomes associated with psychopathology (150), implying a mechanistic pathway relevant for prevention of atypical brain development.
Finally, related to inflammation, from pregnancy, maternal serum immune markers have been shown to play a role in infant brain development (5,(151)(152)(153)(154) by affecting, for example, neurotransmitters (155), glial activity (156), and telomere length (157). Postnatally, one study reported that higher maternal social support at 12 months was associated with decreasing levels of infant salivary C-reactive protein, a proinflammatory marker (158). Inflammation may also modify growth factors in the offspring brain (159). Interestingly, maternal social support after the delivery has recently been reported to positively affect neonatal tooth line width, a biomarker linked with prenatal adversity (160). The associations were hypothesized to occur through insulin-like growth factors (e.g., IGF-1 and IGF-2) (160) that are relevant for not only hard tissue but also neural development (161). Such preliminary evidence suggests that positive environments may reduce inflammatory markers in the infant, with potential to counteract immune-mediated negative brain outcomes postnatally.

Neurobiology of Resilience-Promoting Postnatal Environments
Taken together, findings from this review suggest that altered brain development following exposure to prenatal adversity is less likely when children are provided with resiliencepromoting early postnatal environments. This conclusion is complemented by translational research on animals suggesting that high-quality maternal caregiving and environmental enrichment may reverse the effects of prenatal stress on brain structure and function in offspring (22,32). For instance, prenatal stress leads to decreases in dendritic spine density and volume in the hippocampus, while postnatal environmental enrichment and high-quality care lead to increases in these measures of hippocampal morphology (32,63). Similar proximal postnatal environmental factors (parental sensitivity, cognitive stimulation) have been associated with promotive influences on brain development in humans (64,66,84,86,98,114). These factors might mitigate the negative effects of prenatal adversity on brain structure and function and neurocognitive outcomes in early childhood (78,83). Furthermore, distal environmental factors, such as higher SES and social support provided to the family, that lead to children receiving high-quality caregiving and cognitive stimulation are also moderators of associations between prenatal adversity and neurocognitive and brain outcomes (111,128,131).
Considering distal and proximal factors together, one interpretation of the extant findings is that while prenatal stress may lead to smaller hippocampal volumes (33), higher SES and higher caregiving quality may offset these effects by contributing to larger hippocampal volumes over time (78,114). Although based on fewer studies, a similar pattern of findings has been described for amygdala morphology and frontolimbic connectivity (84,111). At a more fine-grained level, these postnatal promotive and protective effects on the infant brain may occur through, for example, epigenetic and immunerelated mechanisms that play a role in modifying HPA axis functioning (150). However, direct neuroscientific evidence of postnatal resilience processes for human infant brain development after exposure to prenatal stress is still limited, and very few published studies of these associations in humans have been equipped to make causal inferences.
The field is at an early stage of understanding the neural mechanisms underlying postnatal buffering of prenatal adversity effects on emotional, neurocognitive, and behavioral health in children (47)(48)(49)(50)(51)(52)(53)(54)(55)57). Research points to specific neural mechanisms that are plausibly related to these outcomes. For emotional outcomes, frontolimbic circuitry has been associated with stress and emotion processing and regulation (40,41). For neurocognitive outcomes, the hippocampus and amygdala have been associated with learning and memory (42). Therefore, positive early postnatal environments may counteract the negative effects of prenatal adversity exposure on learning and emotion regulation, leading to a greater likelihood of better emotional and behavioral health.

Limitations and Future Directions
The major limitations of this research include variability in measurements of prenatal stress, limited examination of brain outcomes, and lack of conceptual clarity around resilience processes. While we focused on maternal psychosocial stress, prenatal stress is a broad term used to refer to various factors that can affect the intrauterine environment (162). Maternal cortisol measures and child low birth weight and preterm birth were included as early indicators of prenatal stress-related adversity. These measures reflect different exposures and may have differential effects on the developing brain. In addition, there are only a limited number of studies examining resilience processes that may alter associations between prenatal stress and infant brain outcomes. Therefore, we included studies of neurocognitive and HPA axis outcomes, which allow reasonable insights into moderating processes at the neural level. Finally, we acknowledge the complexities associated with resilience research in general, including discerning between protective and promotive factors (25). To date, hypotheses have been typically insufficient in terms of the type of interactions being tested. It is important for future research to specify the outcomes examined and the moderation model being tested.
Measures of brain structure and function are sensitive indices that can reveal mechanistic effects, even when differences in observable neurocognitive and mental health outcomes are not yet detectable (163). More research using neuroimaging techniques is needed to examine the extent to which postnatal environments may attenuate or reverse the negative effects of prenatal adversity exposure on the infant brain. One challenge may be that interaction testing requires sufficient statistical power, and most neuroimaging studies to date have relied on small sample sizes (164), reducing the ability to detect significant interactions. In addition, existing correlational data could be systematically gathered and analyzed together (e.g., meta-analyses, ideally with individual-level infant data) to ensure statistical power for detecting moderation effects. Future longitudinal research studies that include serial magnetic resonance imaging scans will allow investigating how supportive postnatal environments may modify brain developmental trajectories initiated by prenatal stress exposures. Ongoing multisite studies, such as the HBCD (HEALthy Brain and Child Development) Study (165), will also provide opportunities to address the questions discussed here. Finally, given evidence from RCTs regarding the positive influences of caregiving and social resources on child brain in preterm infants (72,86,125), future studies should use RCT designs while focusing specifically on prenatal stress.
There are also other possible directions for future research. For example, nonhuman animal research relies heavily on exposures such as physical stimulation by the mother (e.g., licking). Human research has also provided promising findings on the role of stroking or skin-to-skin contact for infant brain, especially in preterm infants (49,(84)(85)(86), but such research in the context of prenatal stress is scarce. In line with bioecological models (44), future research could also consider approaches that integrate multiple levels of environmental influences and caregivers (e.g., the family village model) (166).

Implications for Interventions
Universal mental health preventive prenatal interventions have been proposed to provide screening and care to all women during pregnancy (5). Our review suggests that such early intervention protocols should including both prenatal and postnatal components focusing on ensuring high-quality early postnatal environments for the newborn and social or socioeconomic support for the family. For instance, there is evidence for the efficacy of programs targeted to enhance caregiving quality, even in high-risk populations (56,57,167), with positive effects on child brain outcomes and methylation of gene pathways related to neural development (61,70,86) and ultimately, child neurocognitive and behavioral development (57). Interventions targeting coparenting and social support (couple-focused prevention) may also associate with adaptive child development (168). Indeed, interventions can be targeted at distal and proximal environmental factors to benefit children and families. As an important illustration, RCT evidence suggests that ensuring the socioeconomic stability of families during periods of heightened neural plasticity supports child brain development (125). Neurobiological measures should continue to be integrated into the evaluation of programs that provide support across the perinatal period (e.g., nurse-family partnerships) to fully understand how they benefit children.

CONCLUSIONS
Elevated maternal stress during pregnancy has been found to affect infant brain development and lead to a higher risk for mental health problems in offspring. During the early postnatal period, a window of heightened malleability to environmental influences, resilience-promoting environmental factors may reduce the effects of prenatal stress exposure on infant brain development. In an era of research that increasingly emphasizes the role of resilience in the prevention of psychopathology (24), positive environmental factors during early postnatal life should be considered alongside prenatal adversity exposures as crucial contributors to long-term mental health outcomes.