(a) Motor development is embodied: Opportunities for action depend on the currentstatus of the body. (b) Motor development is embedded: Variations in the environment create andconstrain possibilities for action. (c) Motor development is enculturated: Social and culturalinfluences shape motor behaviors. (d) Motor development is enabling: New motor skills createnew opportunities for exploration and learning that instigate cascades of development acrossdiverse psychological domains. Fo

Intense maternal stress, such as exposure to HurricaneKatrina, is associated with low birth weight (,2.5 kg) (51).However, these effects are independent of maternal mentalhealth, further underscoring the distinct influences of spe-cific forms of maternal adversity. An extensive review (52)reveals an influence of severe stress (e.g., death of a spouse) onoffspring birth weight, as well as of factors such as socialsupport that moderate the impact of stress, but no consistentevidence for the influence of maternal anxiety or depression(also see reference 53). The exceptions are studies showingan association between “pregnancy-associated anxiety” andbirth outcomes, which again underscores the specificity ofdifferent forms of maternal “adversity.”

Studies employing momentary assessments ofmaternal mood states show covariation between negativemood in pregnancy and salivary cortisol (e.g., 43); however, anumber of studies find little or no association between maternalcortisol levels and measures of maternal stress, anxiety, or de-pression (25, 44–47). Indeed, pregnancy in humans and othermammals is associated with dampened HPA stress reactivity.Detailed studies reveal no association of maternal salivary (19),plasma, or amniotic cortisol levels (46, 47) with either maternalstress or anxiety. Sarkar et al. (47) reported a weak correlationbetween maternal anxiety and plasma cortisol, and only in earlypregnancy. In contrast, O’Connor et al. (48) reported an asso-ciation between maternal depression and diurnal cortisol levelsin a low-socioeconomic-status sample, while one study (49)has reported increased cortisol levels in pregnant womenwith comorbid anxiety and depression. These findings sug-gest that the relation between maternal mental health andglucocorticoid exposure may lie within specific subgroups,including women with more severe mental health conditions

And subtle distinctions within specific forms of“adversity” seem important. For example, while trait anxietyaccounts for a proportion of pregnancy-related anxiety,pregnancy-related anxiety may have greater predictive valuefor obstetric or child outcomes than more global measures ofmaternal anxiety (e.g., 41, 42)

The Term “Maternal Adversity

Studies of 11b-HSD-2 null mice provide ev-idence for a causal association between 11b-HSD-2, reducedbirth weight, and anxiety-like behavior in adulthood

A deficiency of 11b-HSD-2 leads tooverexposure of the fetus to cortisol and lower birth weight

While lipophilic steroids easily crossthe placenta, fetal glucocorticoid levels are much lower thanlevels in maternal circulation (36) because of placental 11b-hydroxysteroid dehydrogenase type 2 (11b-HSD-2), whichconverts active glucocorticoids (cortisol and corticosterone)to inert 11-keto forms (cortisone, 11-dehydrocorticosterone)(34, 35).

Cortisol levels in cord blood are increased inintrauterine growth retardation, implicating endogenouscortisol in fetal growth

late pregnancy, when fetal growth isaccelerated.

Antenatal treatment with the highly catabolic glucocorticoidsreduces birth weight

Recent studies thatovercome this complication by imaging shortly after birth(22, 33) reveal that antenatal maternal anxiety predicts var-iation in microstructure of regions important for cognitive-emotional function (right insula and dorsolateral prefrontalcortex), sensory processing (right middle occipital cortex),and socioemotional function (right angular gyrus, uncinatefasciculus, posterior cingulate, and parahippocampus). Theseright lateralized clusters predict infant internalizing behavior

An issue for studies of fetal neurodevelopment is thatoutcome measures are often collected well after birth. Studiesof fetal development using measures of cardiac function oractivity reveal clear effects of maternal emotional well-beingon fetal physiology (31, 32). While imaging studies of olderchildren suggest prenatal influences on brain structure (e.g.,30), there remains the possibility that these effects areconfounded by postnatal experience.

An-tenatal maternal anxiety is associated with gray matter vol-ume reductions in the prefrontal cortex, the medial temporallobe, the lateral temporal cortex, and the postcentral gyrus aswell as the cerebellum extending to the middle occipital gyrusand the fusiform gyrus (30)

Cortical thinning is a proposed endophenotypefor depression and mediates, in part, the relationship betweenantenatal maternal mood and externalizing behaviors.

Antenatal maternal depressive symptoms are also associatedwith cortical thinning, primarily in the right frontal lobes inchildren

The impact of antenatal maternal stress on neurode-velopmental outcomes is well established in rodent andnonhuman primate models (reviewed in references 24, 25).These studies benefit from controls for postnatal influences,including cross-fostering at birth to nonstressed mothers

Finally, community samples revealassociations between self-reported symptoms of depressionor anxiety and neurodevelopment suggesting that effects ofantenatal maternal mood on child outcome occur across thespectrum of symptom severity (14–16, 23) and are not uniqueto clinical samples.

Antenatalmaternal symptoms of anxiety or depression are associatedwith more difficult/reactive infant temperament (17–19),independent of postnatal maternal mental health (18, 19)

paucity

Only a few studies have examined neural structure in relation tobirth weight across the normal range.

Detailed information on fetal growth, such as that derivedfrom serial ultrasound scans, is difficult to obtain. Epide-miological studies instead use proxy measures, such as birthweight, gestational age, and, less often, birth length, headcircumference, placental weight, or ponderal index (4). Wefocus here on birth weight, which is the most commonlystudied measure of fetal development, but it is important tonote that birth weight, gestational age, and birth length mayreflect different underlying mechanisms with independenteffects on specific mental health outcomes

The association between birth weight and cardiometabolichealth catalyzed interest in the fetal origins of non-communicable disease (1, 2). Importantly, the relation be-tween birth weight and metabolic health is broadlycontinuous: the risk of metabolic disease declines with in-creasing birth weight up to the point of macrosomia, whereit once again increases. More successful pregnancies (i.e.,optimal fetal growth) are associated with better metabolichealth. These studies (1, 2) contributed to the emphasis onmaternal health as a global priority for the World HealthOrganization (http://www.who.int/pmnch) and the Interna-tional Monetary Fund (http://www.imf.org/external/np/exr/facts/mdg.htm)

The quality of fetal growth and development predicts the riskfor a range of noncommunicable, chronic illnesses. Theseobservations form the basis of the “developmental origins ofhealth and disease” hypothesis, which suggests that the in-trauterine signals that compromise fetal growth also act to“program” tissue differentiation in a manner that predisposesto later illness. Fetal growth also predicts the risk for laterpsychopathology. These findings parallel studies showingthat antenatal maternal emotional well-being likewise pre-dicts the risk for later psychopathology. Taken together, thesefindings form the basis for integrative models of fetal neu-rodevelopment, which propose that antenatal maternaladversity operates through the biological pathways associ-ated with fetal growth to program neurodevelopment. Theauthors review the literature and find little support for suchintegrated models. Maternal anxiety, depression, and stressall influence neurodevelopment but show modest, weak, orno associations with known stress mediators (e.g., gluco-corticoids) or with fetal growth. Rather, compromised fetaldevelopment appears to establish a “meta-plastic” state thatincreases sensitivity to postnatal influences. There also re-main serious concerns that observational studies associatingeither fetal growth or maternal mental health with neuro-developmental outcomes fail to account for underlyinggenetic factors. Finally, while the observed relation betweenfetal growth and adult health has garnered considerableattention, the clinical relevance of these associations remainsto be determined. There are both considerable promise andimportant challenges for future studies of the fetal origins ofmental health

Taken together, thesefindings form the basis for integrative models of fetal neu-rodevelopment, which propose that antenatal maternaladversity operates through the biological pathways associ-ated with fetal growth to program neurodevelopment.

hese findings parallel studies showingthat antenatal maternal emotional well-being likewise pre-dicts the risk for later psychopathology.