Prenatal and Postnatal Maternal Distress and Offspring Temperament: A Longitudinal Study

Abstract

Maternal distress experienced prenatally and in the child’s first year of life has been associated consistently with offspring psychopathology. Less research has addressed whether it is also associated with variations in psychological traits. The present research used two samples from the Longitudinal Study of Australian Children to examine the association between maternal prenatal and postnatal distress and offspring temperament across childhood. Maternal distress experienced by mothers was associated with higher offspring reactivity and lower offspring persistence. These associations replicated across both types of maternal distress (prenatal/postnatal), across two different measures of temperament, and remained significant controlling for maternal distress concurrent with the temperament measures and controlling for maternal personality. There was less evidence that either type of maternal distress was associated with sociability and no evidence that it was associated with the trajectory of the three dimensions of temperament across childhood. Maternal distress is associated with traits that reflect dysregulation and may be one mechanism through which prenatal and early life factors contribute to individual differences in psychological function.

The burden of maternal depression for offspring psychopathology is well documented (Goodman et al., 2011). Mothers who experienced depression during their child’s prenatal development have offspring who are at greater risk of internalizing and externalizing symptoms across childhood (Gentile, 2017) and major depression during adolescence (Pearson et al., 2013) and adulthood (Plant et al., 2015). Similar patterns are apparent for maternal depression early in the offspring’s life: Offspring of mothers who experienced depression and other forms of distress are likewise at greater risk of symptomatology (Conroy et al., 2012; Field, 2018) and depression in adolescence (Murray et al., 2011) and adulthood (Weissman et al., 2016).

The association between maternal depression and offspring psychological outcomes likely extends past psychopathology to normal variation in psychological function. In particular, both prenatal (Madigan et al., 2018) and postnatal (Hanington et al., 2010) maternal depression have been associated with offspring personality in early childhood. Personality traits are relatively enduring characteristic ways of thinking, feeling, and behaving (McCrae and John, 1992) that emerge early in life and develop across the lifespan (Shiner, 2015). In adulthood, trait psychological function is typically operationalized along five dimensions (Costa and McCrae, 1992): Neuroticism (the tendency to experience negative emotions), extraversion (the tendency to be social and experience positive affect), openness (the tendency to be creative), agreeableness (the tendency to be trusting), and conscientiousness (the tendency to be disciplined). In childhood, personality traits are conceptualized as temperament, and theoretical models of temperament tend to focus on traits related to neuroticism (e.g., negative affectivity), extraversion (e.g., positive affectivity), and conscientiousness (e.g., effortful control), although some models also consider aspects of agreeableness in their conceptualizations (see Shiner and DeYoung, 2013 for a review). Aspects of temperament have been linked to FFM personality traits, particularly the dimension of reactivity has been linked to neuroticism, sociability has been linked to extraversion, and persistence has been linked to conscientiousness (De Pauw and Mervielde, 2010).

Maternal depression is associated most commonly with traits related to negative affectivity (a component of neuroticism) in the first year of life. Depressive symptoms experienced during pregnancy, for example, are associated with higher offspring temperamental negativity assessed at 6 weeks (Luecken et al., 2015), with higher negative affect at three months (Rouse and Goodman, 2014), and with higher negative affect at assessments across three, six, and 12 months (Lusby et al., 2016). These associations further persist across childhood: A meta-analysis found that prenatal depression and anxiety were associated with negative affect in samples that ranged from infancy to adolescence (Madigan et al., 2018). Similar associations are also apparent for maternal depression experienced during offspring childhood. Maternal depression, for example, is associated with greater offspring negative affect (Goodman et al., 2011) and higher average neuroticism during childhood (Allen et al., 2019).

Less research has addressed how maternal depression is associated with traits other than negative affectivity. There is some evidence, however, that maternal depression is associated with lower conscientiousness and lower extraversion. Parent depression (both maternal and paternal) in early childhood, for example, is associated with lower inhibitory control in offspring (Gagne et al., 2013), a trait associated with conscientiousness. In addition, maternal major depression during the offspring’s first year of life has been associated with less of an increase in conscientiousness from three to nine years of age, although it was not related to average level of conscientiousness (Allen et al., 2019). A meta-analysis found that maternal depression was associated with less positive affect, one component of extraversion, in childhood (Goodman et al., 2011). And, more generally, maternal depression has been associated with a difficult temperament: Depressive symptoms and distress during pregnancy are related to more difficult temperament in the first year of life (Kerstis et al., 2013) and maternal depressive symptoms experienced both prenatally and at offspring four years of age are associated with more emotional and behavioral difficulties at age four (Woolhouse et al., 2016).

As such, although there is consistent evidence that maternal distress is associated with negative affectivity and difficult temperament, fewer studies have addressed how it is associated with other major dimensions of personality. In addition, few studies have measured personality-related traits at multiple time points across childhood. Thus far, most research on maternal depression and offspring psychological traits has relied on an assessment at one point in time (e.g., Gagne et al., 2013; Rouse and Goodman, 2014) or multiple times only in infancy (e.g., Lusby et al., 2016); but see (Allen et al., 2019) for a notable exception for measurement across early and middle childhood). Personality, however, is not static but develops across childhood into adolescence (Soto and Tackett, 2015). There is also variation across childhood that indicates that some children change more or less than the average. Maternal distress may be associated with this variability. The associations may persist into adolescence, as is the case for psychopathology (Murray et al., 2011; Pearson et al., 2013).

The present study uses two relatively large cohorts with up to six assessments of temperament across childhood to examine whether maternal psychological distress, experienced either during pregnancy or the first year of the offspring’s life, is associated with the mean level and trajectory of three traits: Reactivity, sociability, and persistence, components of neuroticism, extraversion, and conscientiousness, respectively. We further examine whether the associations are independent of maternal psychological distress measured concurrent with the traits and whether these associations vary across offspring gender. Finally, we examine whether these associations are independent of mothers’ personality traits in one cohort with relevant data. We expect that maternal distress, experienced either during and after pregnancy, will be associated with higher offspring Reactivity and lower offspring Persistence and Sociability.

Method

Participants and Procedure

Participants were from the two cohorts of the Longitudinal Study of Australian Children (LSAC; Sanson et al., 2002). LSAC was initiated in 2003 to study the development of children and their families from across Australia. LSAC includes a younger (B) cohort with families who were recruited into the study when the study child was an infant (0–1 years of age) and an older (K) cohort with families that who recruited into the study when the study child was 4 or 5 years old. Families are re-interviewed every two years. As part of the regular assessment, primary caregivers reported on the study child’s temperament. Measures of temperament were included in every assessment up through when the study child was 14–15 in the older cohort; caregivers in the younger cohort still provide ratings of temperament (see below). At the first wave of the study, mothers reported on their experience of psychological distress during pregnancy (younger cohort) or during the first year of the child’s life (older cohort). In both cohorts, mothers also provided ratings of their psychological distress concurrent with the temperament ratings in each wave. Finally, for the older cohort only, mothers reported on their own personality traits once, when the children were 16–17. See Figure 1 for an overview of the measures used across the seven waves of LSAC for each cohort.

Figure 1.

Timeline of measures in the Longitudinal Study of Australian Children (LSAC) for the younger cohort (A) and older cohort (B). SATC=Short Temperament Scale for Children. SATI= School Age Temperament Inventory. K6=K6 Distress Scale (measure of maternal psychological distress). BFI-10=Big Five Inventory (measure of maternal personality). Prenatal distress and postnatal distress were measured at Wave 1 in the younger and older cohorts, respectively.

Families were included in the analytic samples if information was available on mother’s psychological distress either during pregnancy (younger cohort) or in the child’s first year of life (older cohort), ratings of temperament were available at least once over the study period, and the relevant covariates were available (see below). A total of 3,735 families met these criteria in the younger cohort and 3,714 families met these criteria in the older cohort.

Measures

Offspring temperament.

The temperament data were collected at up to six time points when study children were between the ages of 4 and 15 years old. Two measures of temperament were used depending on the age of the study child: Temperament was assessed using items from the Short Temperament Scale for Children (STSC; Sanson et al., 1994) from ages 4 to 7 and with the School Age Temperament Inventory (SATI; McClowery, 1995) from ages 8 to 15. The two measures were used to ensure age-appropriate items that measured temperament. The STSC and the SATI both measured three aspects of temperament: Sociability, reactivity, and persistence. Example items for sociability were “This child will go up to strange children and join in their play” and “Approaches children his/her age even when he/she doesn’t know them”, respectively, for the STSC and the SATI; example items for reactivity were, “When this child is angry about something, it is difficult to sidetrack him/her” and “When angry, yells or snaps at others”, respectively, for the STSC and the SATI; example items for persistence were, “This child stays with an activity for a long time” and “Goes back to the task at hand after an interruption”, respectively, for the STSC and the SATI. Each item was rated on a scale from 1 (almost never) to 6 (almost always) on the STSC and from 1 (never) to 5 (always) on the SATI. Items were reverse scored when necessary and the mean taken in the direction of the trait label.

Maternal distress.

Psychological distress during pregnancy was measured in the younger cohort with the item, “During this pregnancy, did you have problems with stress, anxiety or depression?” at the first assessment when the study child was an infant. Psychological distress during the postnatal period was measured in the older cohort with the item, “After the birth of the study child, did you suffer from Post-Natal Depression?” at the first assessment when the study child was about four years old. In both cohorts, participants responded either yes (=1) or no (= 0). In addition, current psychological distress was measured with the K6 Distress Scale (Kessler et al., 2003) in both cohorts. The K6 Distress scale included items on the frequency in the past 4 weeks of feeling worthless, nervous, hopeless, restless or fidgety, that everything is an effort, and so sad that nothing could cheer you up, rated on a scale from 1 (none of the time) to 5 (all of the time).

Maternal personality.

Maternal personality was measured with the Big Five Inventory-10 (BFI-10), a brief but validated measure of five factor model personality traits (Rammstedt and John, 2007). Each trait was measured with two items (e.g., “gets nervous easily” and “relaxed, handles stress well [reverse scored]” for neuroticism) on a scale from 1 (disagree strongly) to 5 (agree strongly). Items were reverse scored when necessary in the direction of the trait label and the mean taken across the two items.

Covariates.

Covariates included child gender (male/female) and indigenous status (yes/no), and the mother’s age at when the child was born, child age at baseline, whether the household had one parent (yes/no), parent education (highest years of education for either parent), parent income (highest weekly income for either parent, log transformed), and neighborhood SES (Socio-Economic Indexes for Areas: Index of Relative Socio-economic Disadvantage; Pink, 2006).

Analytic Strategy

We used Hierarchical Linear Models (HLM; Raudenbush and Bryk, 2002) to examine the association between prenatal or postnatal maternal distress and child temperament and its longitudinal development from 4–15 years old. The three traits were assessed with different measures at different ages (the STSC from ages 4–7 and the SATI from ages 8 and up) and maternal distress was assessed with different measures in the younger and older cohorts. Therefore, separate HLMs were calculated for each of these groups. At level-1, a linear model was fit for each family predicting child temperament at each timepoint from mean-centered child age. At level-2, the intercept and slope of the level-1 models for each child were predicted by prenatal or postnatal maternal distress, controlling for covariates. All level-2 predictors were mean-centered and standardized. Temperament outcomes were left in the original units.

Results

Descriptive statistics for all study variables are shown in Table 1. In the younger cohort, approximately 18% of mothers reported having experienced stress, anxiety or depression during their pregnancy with the study child. In the older cohort, approximately 15% of mothers reported experiencing postnatal depression. The average score on the K6 distress scale was 1.52 (SD = 0.58) on a 1–5 scale across the up to six waves of assessments.

Table 1.

Descriptive Statistics for All Study Predictors

Predictors	Mean (Standard Deviation) or % (n)
	Younger Cohort	Older Cohort
Study child
Age at baseline (years)	4.77(0.22)	4.83(0.24)
Gender (female)	48.57% (1,814)	48.41% (1,798)
Indigenous (yes)	3.02% (113)	2.81% (104)
Mother’s age at birth (year)	29.97(5.10)	30.76(5.09)
Highest parent education	15.42(2.44)	15.85(2.32)
Parent income (log transformed)	2.93(0.29)	3.07(0.31)
Neighborhood SES (SEIFA)	1,012(56.91)	1,017(62.26)
Maternal Distress
Prenatal distress	18.07% (675)	--
Postnatal distress	--	15.27% (567)
Distress concurrent with temperament	1.50(0.55)	1.54(0.59)
Maternal Personality
Neuroticism	--	2.71(1.00)
Extraversion	--	3.36(1.04)
Conscientiousness	--	4.10(0.80)

Note. N=3,735 for the younger cohort and N=3,714 for the older cohort. Values represent a mean (standard deviation) or percent (n). SEIFA = Socio-Economic Indexes for Areas (Pink, 2006).

Prenatal and postnatal maternal distress were associated with offspring personality (Table 2). In both cohorts, greater maternal distress was associated with the intercept of reactivity and persistence: Mothers who reported either prenatal or postnatal distress had offspring who scored higher in reactivity and lower in persistence. Additionally, in the older cohort, mothers who reported postnatal distress had offspring who scored lower in sociability. These associations were apparent on both measures of temperament for reactivity and persistence, but sociability was only significant on the SATI measure used when the children were older. Distress was unrelated to the trajectory of any of these traits across childhood. As such, maternal distress was associated with average levels of reactivity and persistence, but not with change over time.

Table 2.

Association Between Prenatal (Younger Cohort) and Postnatal (Older Cohort) Maternal Distress and the Trajectory of Temperament

Outcome	Predictor	STSC, 4–7yo	SATI, 8–15yo	SATI, 8–15yo, Controlling for Concurrent Maternal Distress
		Prenatal Distress (Younger Cohort)
Sociability	Intercept	3.975[3.942, 4.007], <.001	3.401[3.378, 3.423], <.001	3.397[3.374, 3.419], <.001
	Main Effect	0.005[−0.028, 0.038], .778	−0.019[−0.042, 0.004], .100	−0.017[−0.041, 0.006], .139
	Age	0.152[0.137, 0.167], <.001	−0.011[−0.017, −0.006], <.001	−0.011[−0.017, −0.005], <.001
	Interaction	−0.005[−0.020, 0.010], .522	−0.003[−0.008, 0.003], .390	−0.002[−0.008, 0.004], .593
Reactivity	Intercept	2.450[2.425, 2.475], <.001	2.459[2.437, 2.482], <.001	2.455[2.433, 2.478], <.001
	Main Effect	0.086[0.061, 0.111], <.001	0.072[0.049, 0.094], <.001	0.057[0.034, 0.079], <.001
	Age	−0.118[−0.131, −0.104], <.001	−0.035[−0.041, −0.029], <.001	−0.037[−0.043, −0.031], <.001
	Interaction	−0.004[−0.018, 0.009], .556	0.003[−0.003, 0.009], .295	0.003[−0.003, 0.009], .315
Persistence	Intercept	3.957[3.931, 3.983], <.001	3.445[3.422, 3.469], <.001	3.449[3.425, 3.473], <.001
	Main Effect	−0.039[−0.065, −0.013], .003	−0.068[−0.092, −0.045], <.001	−0.055[−0.079, −0.031], <.001
	Age	0.066[0.052, 0.079], <.001	0.057[0.050, 0.063], <.001	0.058[0.051, 0.065], <.001
	Interaction	−0.009[−0.023, 0.005], .201	0.002[−0.005, 0.009], .523	0.002[−0.005, 0.009], .654
		Postnatal Distress (Older Cohort)
Sociability	Intercept	3.931[3.896, 3.967], <.001	3.365[3.343, 3.387], <.001	3.367[3.344, 3.389], <.001
	Main Effect	−0.010[−0.046, 0.026], .591	−0.023[−0.045, −0.000], .047	−0.016[−0.039, 0.008], .188
	Age	0.091[0.075, 0.107], <.001	−0.019[−0.024, −0.015], <.001	−0.020[−0.025, −0.016], <.001
	Interaction	0.004[−0.013, 0.020], .658	−0.004[−0.009, 0.000], .080	−0.004[−0.009, 0.001], .086
Reactivity	Intercept	2.550[2.524, 2.576], <.001	2.371[2.349, 2.394], <.001	2.370[2.348, 2.393], <.001
	Main Effect	0.062[0.036, 0.088], <.001	0.060[0.037, 0.082], <.001	0.039[0.016, 0.062], .001
	Age	−0.136[−0.150, −0.122], <.001	−0.022[−0.027, −0.018], <.001	−0.020[−0.025, −0.016], <.001
	Interaction	0.004[−0.010, 0.019], .553	−0.002[−0.007, 0.002], .273	−0.003[−0.007, 0.002], .267
Persistence	Intercept	3.986[3.959, 4.013], <.001	3.528[3.505, 3.552], <.001	3.528[3.505, 3.552], <.001
	Main Effect	−0.036[−0.063, −0.009], .010	−0.066[−0.090, −0.043], <.001	−0.055[−0.079, −0.031], <.001
	Age	0.049[0.034, 0.063], <.001	0.042[0.037, 0.047], <.001	0.042[0.037, 0.047], <.001
	Interaction	−0.001[−0.016, 0.014], .897	0.002[−0.003, 0.007], .493	0.001[−0.004, 0.006], .667

Note. Values are regression coefficients [95% confidence intervals], p-values. All predictors were standardized, outcomes are in original units. STSC = Short Temperament Scale for Children (Sanson et al., 1994). SATI = School-Age Temperament Inventory (McClowery, 1995). Bolded values are statistically significant main effects and interactions, p<.05.

We did several sensitivity analyses to evaluate the robustness of the associations. First, most mothers self-reported symptoms of psychological distress concurrent with at least some ratings of child’s temperament (younger cohort N = 3,445; older cohort N = 3,397); we thus examined the inclusion of concurrent distress as a time-varying covariate. Note that we could only include this time-varying covariate for the SATI because there were not enough measurement occasions with the STSC (i.e., the SATI measure had up to four time points, whereas the STSC had only two time points). The pattern of associations was similar when controlling or not controlling for concurrent psychological distress (Table 2), with the one exception that the association between post-natal depression and average level of sociability in the older cohort was no longer significant after controlling for concurrent distress. In addition, we examined the associations between concurrent maternal distress and child temperament, controlling for prenatal or postnatal distress. In both the younger and older cohort, greater maternal psychological distress was associated with lower child sociability and persistence as well as higher child reactivity (Table 3).

Table 3.

Association Between Concurrent Maternal Distress and the Trajectory of Temperament (8–15yo only)

Cohort	Outcome	Value
Younger Cohort	Sociability	−.034[−.049, −.019], <.001
	Reactivity	.106[.089, .123], <.001
	Persistence	−.087[−.105, −.068], <.001
Older Cohort	Sociability	−.048[−.063, −.032], <.001
	Reactivity	.100[.084, .116], <.001
	Persistence	−.075[−.092, −.057], <.001

Note. Values represent a regression coefficient [95% confidence interval], p-value. All predictors were standardized, outcomes are in original units. Bolded values are statistically significant, p<.05.

Second, a subset of mothers in the older cohort (N = 2,334) reported on their own personality when their child was 16–17. We included the analogous trait as an additional covariate in each analysis. In this subsample, the pattern of associations was unchanged when controlling or not controlling for maternal personality (Supplemental Table S1). Though, unlike with the full sample, postnatal depression was not significantly associated with persistence on the STSC measure used when the children were younger in this subsample. Finally, with the full sample in both cohorts, moderation tests indicated that none of the associations between prenatal or postnatal maternal distress and child temperament were moderated by offspring gender (all interactions ns).

Discussion

In two large cohorts with longitudinal assessments of temperament, maternal distress had clear and replicable associations with personality: Prenatal and postnatal maternal distress were associated with higher offspring reactivity and lower offspring persistence. The association with offspring sociability was more tenuous. Further, neither aspect of maternal distress was associated with the trajectory of the three traits across childhood, which indicates that the associations with temperament did not strengthen or dissipate over time. Further, the associations were robust and mostly unchanged after adjusting for concurrent maternal psychological distress and maternal personality.

Maternal Distress and Temperament

It is common for women to experience psychological distress during pregnancy and in the first year of their child’s life: It is estimated that 20% of women experience prenatal depression and approximately 12% of women experience postnatal depression (Woody et al., 2017). The percentages in this study were roughly similar to these estimates (18% and 15%, respectively). There are well-established associations between these experiences of distress and risk for psychopathology among offspring (Gentile, 2017; Plant et al., 2015). There is also growing literature that indicates that the associations extend past psychopathology to normal variations in psychological functioning (Madigan et al., 2018). Much of this literature has focused on infancy (e.g., Lusby et al., 2016), but, similar to psychopathology (Pearson et al., 2013), there is evidence that it extends at least into middle childhood (Allen et al., 2019).

Consistent with research on psychopathology (Gentile, 2017; Plant et al., 2015) and trait negative emotionality (Madigan et al., 2018), maternal distress was associated with higher offspring reactivity. Reactivity refers to how a child handles anger and responds to frustration (McClowery, 1995; Sanson et al., 1994). This behavioral pattern is one component of trait neuroticism (Costa and McCrae, 1992), which is a risk factor for psychopathology (American Psychiatric Association, 2013). And, in fact, Allen and colleagues (Allen et al., 2019) demonstrate that the association between maternal distress and offspring symptomatology may be mediated through offspring neuroticism. This pattern suggests that distress may be transmitted from mother to child through a trait tendency toward negative emotionality. In this study, accounting for mothers’ neuroticism attenuated the main effect on reactivity by about 30%.

There was also evidence that maternal distress is associated with lower offspring persistence. Persistence refers to how long a child will stay with a task (McClowery, 1995; Sanson et al., 1994). Persistence is a component of conscientiousness, which is critical for better outcomes, including higher education (Shiner et al., 2003) and better health into adulthood (Hampson et al., 2013). Fewer studies have examined the association between maternal distress and aspects of conscientiousness, compared to neuroticism, but there is some evidence that it is associated with lower inhibitory control (Gagne et al., 2013) and with less of an increase in conscientiousness across childhood (but not the mean level; Allen et al., 2019). More research is needed to determine whether maternal distress is associated more strongly with specific facets of conscientiousness more than others.

The evidence for an association between maternal distress and offspring sociability was more nuanced. Sociability is one major component of extraversion; positive emotionality is another major component. Previous research has indicated that maternal distress is associated with lower offspring positive affect (Goodman et al., 2011) but is unrelated to extraversion (Allen et al., 2019). As such, maternal distress may inhibit positive emotionality but its null association with sociability reduces its overall association with extraversion. It is important to note, however, that among the older children, postnatal distress was associated with lower sociability and that the association with prenatal distress was in the same direction but not significant. In contrast, both measures of distress were virtually unrelated to sociability measured in early childhood. As such, it may be that maternal distress is associated with lower sociability later in childhood. There also may be other environmental factors that are more important for childhood sociability (e.g., neighborhood SES; (Strickhouser and Sutin, 2020).

Potential Mechanisms

There are several mechanisms through which maternal depression and distress may contribute to offspring personality. First, the association may be due to shared genetics. It may be that a genetic predisposition for experiencing distress is also responsible for general tendencies toward neuroticism and conscientiousness. For example, large genome-wide association studies show that variants linked to neuroticism are also predictors of major depressive disorder (de Moor et al., 2015; Okbay et al., 2016). Given that genes can influence multiple phenotypes (i.e., pleiotropy), it is possible that genetic transmission is responsible for the observed associations. Genetically informed designs tend to show a role for both genetics and the environment for offspring psychopathology (Kerr et al., 2013). In a sample of offspring who were adopted, for example, the biological mother’s psychopathology and the adoptive mother’s depressive symptoms were both associated with the offspring’s internalizing and externalizing behaviors in early childhood (Kerr et al., 2013). There is also evidence that genetics may account for most of the association between prenatal maternal distress and offspring symptomatology in early childhood but that environmental exposures account for some of the association later in childhood (Hannigan et al., 2018). It may also be the case that maternal depression is a key mechanism through which a genetic predisposition shapes the psychological functioning of offspring.

Second, there are other biological mechanisms that may contribute to the association. That is, there may be a biological pathway from maternal depression to offspring psychological development through gestational development. For example, maternal inflammatory cytokines have been found to mediate the association between maternal prenatal depression and infant negative affect at six months (Gustafsson et al., 2018). Likewise, greater gene expression in the placenta mediates the relation between maternal prenatal depressive symptoms and offspring negative affectivity at six months (Zhang et al 2018). The present research suggest that this pathway is unlikely to be the sole mechanism because we found similar associations for postnatal distress as for prenatal distress.

Third, there are environmental and behavioral pathways between maternal distress and offspring personality. Maternal depression and depressive symptoms in the first year of the offspring’s life, for example, are associated with less maternal sensitivity (Bernard et al., 2018). Mothers suffering from distress may have difficulty recognizing and responding to their infants’ needs. Maternal depression is further associated with neglectful parenting during preschool and elementary school and such parenting mediates the relation between distress and greater offspring behavioral problems (although this association does not extend into adolescence; Mustillo et al., 2011). Mothers suffering from distress may engage in behaviors that are harmful to the child or encourage a way of responding or may not be responsive to the child’s needs. Of note, treatment of maternal depression early in the offspring’s life may be associated with improvements in difficult temperament through reductions in maternal depressive symptoms (Handley et al., 2017).

Implications for Personality Development

The present research contributes to lifespan models of personality trait development. Much of the research on the origins of personality focuses on genetic underpinnings of traits and heritability estimates that suggest a stronger role for genetics than the environment in early childhood (Briley and Tucker-Drob, 2014). More sophisticated behavioral genetics analyses, however, indicate the importance of both shared and non-shared environment for personality traits (Matteson et al., 2013). The current findings join a growing body of evidence for prenatal and early life correlates of personality traits. For example, children from families with more economic means (Lengua, 2006) and families who live in higher SES neighborhoods (Strickhouser and Sutin, 2020) tend to score lower in neuroticism-related traits and higher in conscientiousness-related traits. Further, healthier nutrition in the first year of life (Galler et al., 2013) and having been breastfed (Sutin et al., 2016) are associated with greater offspring emotional stability in adulthood. Environmental exposures, such as maternal smoking during pregnancy (Sutin et al., 2018) and exposure to pollution in childhood (Reuben et al., 2019) also contribute to personality in adulthood. The prenatal and early life environment (broadly defined) in which individuals grow may shape how trait psychological function develops.

More broadly, FFM personality traits are predictors of important outcomes across the lifespan, including educational achievement (Noftle and Robins, 2007), health and well-being (Strickhouser et al., 2017), and longevity (Jokela et al., 2013). Personality traits may be shaped, in part, by the environments in which children grow up, and, in turn, personality may be one way in which environments shape long-term outcomes. Lifespan models of personality traits and health tend to focus on the downstream correlates of trait dispositions (Friedman et al., 2014). An integration of the emerging findings on prenatal and early life environments on trait development and existing models of personality and health will provide a richer theoretical model to better identify the pathways from early life to adult health and well-being through psychological development.

Strengths and Limitations

The present study had several strengths, including two large cohorts, measures of both prenatal and postnatal maternal distress, multiple assessments of personality across childhood, and consideration of potentially confounding factors. There are also limitations to consider when evaluating this research. First, the measures of maternal distress were brief and only included a single item for both prenatal and postnatal distress. Such an approach is both less reliable than longer measures and does not capture the complexity of distress. Future research would benefit from more detailed scales to examine whether aspects of distress (e.g., symptoms of anxiety versus depression) have differential associations with offspring personality. A clinical diagnosis would also be helpful to examine how clinically identified distress is associated with offspring temperament. Second, we did not have measures of prenatal and postnatal distress in the same sample and thus we were unable to test whether there are independent effects or whether postnatal distress mediates the association between prenatal distress and offspring temperament. Third, we did not have measures of paternal distress concurrent with maternal prenatal/postnatal distress. Such information would be important to measure in future research. Finally, we did not have a genetically informed design (e.g., twin samples, molecular genetics) that would help to estimate genetic and environmental contributions of maternal distress to the traits. When such designs are not possible, it has been recommended to include a parent measure of the same trait as a covariate to account for potential genetic transmission (Hart et al., 2021). In the current study, inclusion of the mother’s analogous personality trait did not change the pattern of associations.

Conclusions

Despite these limitations, the present research indicates that maternal distress is associated with higher reactivity and lower persistence and that the associations are consistent across childhood. These associations replicated across prenatal and postnatal maternal distress and across two measures of temperament in young and middle childhood. Further, these associations remained after controlling for concurrent maternal psychological distress and maternal personality. Overall, the results suggest that maternal distress is associated with personality traits that reflect dysregulation. Future research is needed to determine whether treatment of distress during critical periods has a downstream effect on the development of these personality traits.