Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2025 Sep 25;67(5):e70078. doi: 10.1002/dev.70078

The Dyadic Association of Testosterone With Perceived Social Support in Couples Across the Transition to Parenthood

Nicole M Froidevaux 1,2,✉, Jocelyn Lai 3, Shauna G Simon 1, Laurel Benjamin 4, Jennifer A Somers 5, Douglas A Granger 6,7, Dana Shai 8, Jessica L Borelli 1,✉
PMCID: PMC12464473  PMID: 40999715

ABSTRACT

Social support is associated with positive health outcomes and may be protective during sensitive life periods. Importantly, biological factors, including hormones, are associated with psychological and interpersonal functioning, but less is known about the link between biological factors and perceived social support during the transition to parenthood. The current study examined prenatal testosterone levels as a predictor of postpartum perceived social support among first‐time parents. Participants (N = 110 husband–wife dyads) engaged in an ecologically valid laboratory‐based infant simulator task involving an inconsolable doll. Saliva samples were collected prenatally before and after the infant simulator task. Perceived social support was self‐reported prenatally and at 6 months postpartum. On average, fathers’ baseline testosterone levels were higher than mothers’ (p < 0.001). Most fathers and mothers demonstrated an increase in testosterone in response to the parenting task (n fathers = 66; n mothers = 67). An actor–partner interdependence model tested the association of prenatal baseline testosterone and testosterone reactivity predicting postpartum perceived social support, accounting for prenatal perceived social support. Findings revealed that (1) lower mother prenatal baseline testosterone was associated with higher mother postpartum perceived social support and (2) lower father prenatal testosterone reactivity was associated with higher mother postpartum perceived social support. No paths predicted fathers’ postpartum perceived social support. Interpretations and implications of these findings are discussed.

Keywords: couple relationships, perceived social support, testosterone, transition to parenthood

1. Introduction

Social support is imperative for adjustment among first‐time parents during the transition to parenthood. While biological measures of testosterone, a steroid hormone, have been linked to positive paternal behaviors (Edelstein et al. 2017; Endendijk et al. 2016; Gordon et al. 2017) and familial outcomes (Cárdenas et al. 2023; Gettler et al. 2020), its role in predicting psychological factors is unclear. In the current study, we investigated how testosterone levels were associated with perceived social support in the context of a simulated co‐parenting stressor task during the transition to parenthood.

1.1. The Transition to Parenthood

Transitions in romantic relationships denote critical inflection points during which one or both partners experience a substantial shift in life stage or social role. Although transitions in close relationships can bring about positive changes, such as growth and excitement, they also often confer stress. Both partners may have to adjust their roles, ways of interacting with one another, or expectations for the relationship in order to meet the demands of the transition. The transition from romantic partners to co‐parents can present various challenges, including increased strain within the relationship (Theiss et al. 2013), reductions in commitment (Ter Kuile et al. 2021), and increased conflict (see Kluwer [2010] for a review). Indeed, some research also suggests that parents experience greater decline in marital satisfaction over time compared to nonparent couples (Lawrence et al. 2008).

1.1.1. The Importance of Social Support

Given the known vulnerability of romantic relationships during this life phase, social support from one's partner may be one critical ingredient to weathering the transition. For instance, higher perceptions of support from one's partner may aid couples in their adjustment to the challenges posed by this transition—perhaps it helps couples feel like they are navigating this challenge on the same team, rather than handling big challenges alone. Indeed, higher perceived social support is linked to lower distress in couples during the postpartum period (Castle et al. 2008; Gillis et al. 2019). Further, mothers who feel more supported or more satisfied with their partner's support not only adjust better to their transition to parenthood (Tietjen 1985) but also experience higher physical and mental well‐being (Dennis and Ross 2006; Stapleton et al. 2012).

Beyond the context of parenting, higher perceived social support from close others is an important predictor of personal and relational well‐being and is associated with favorable physical and mental health outcomes (Cohen and Wills 1985; Uchino 2009). Social support can buffer the links between stress and poor physical and mental health (see Cohen and Wills [1985] for theory on stress buffering). Within romantic relationships, people who perceive themselves as receiving more support from their partner report higher relationship satisfaction (e.g., Bar‐Shachar et al. 2023; Nawaz et al. 2014).

Although perceived social support has been established as a predictor of adjustment, less research has focused on understanding what predicts social support. Some very early work examined characteristics of the support network in understanding perceived social support and found that frequency of contact with close others and number of close others were both positively associated with perceived social support (Cutrona 1986). More recent work has examined psychological predictors of perceived social support, finding that individuals who are more secure in their attachments (Mikulincer and Shaver 2009; Stanton and Campbell 2014), those who are less depressed (Maher et al. 2006), those who are less stressed (Maneshi Azghandi et al. 2022), and those with specific personality compositions (e.g., people high in extraversion; Swickert et al. 2010) perceive those around them as more supportive. Very little work, however, has examined the role of physiology in perceived social support. Only one study has examined the association of physiological emotion regulation (indexed by higher heart rate variability) with perceived social support from close others (Pourmand et al. 2023), despite the widely documented association between physiological and psychological experiences. In the current study, we expand this work using a longitudinal, dyadic design to assess whether testosterone, a hormone with relevance for parenthood and familial relationships (Endendijk et al. 2016), is a predictor of perceived social support during the transition to parenthood.

1.1.2. The Importance of Testosterone

Testosterone is a steroid hormone produced by the gonads and adrenal glands, circulating through the bloodstream. Generally, males have higher levels than females (see Barry et al. [2011] for a meta‐analytic review), but these levels can vary over a lifetime, with distinct sex differences (Handelsman et al. 2016). Testosterone can be measured at baseline. It can also be measured in response to discrete events, such as laboratory tasks or life events, when it is referred to as testosterone reactivity. There is reason to believe that both baseline testosterone and testosterone reactivity levels may be relevant for couples as they transition to parenthood.

Baseline levels of testosterone can inform how typical hormone levels are associated with psychological outcomes. Testosterone reactivity levels, measured as the difference between baseline levels and immediately following a discrete event (e.g., a parenting stress task), can inform how biological changes in response to stress (prevalent during the transition to parenthood) are associated with psychological outcomes. In the sections that follow, we briefly summarize the state of literature and gaps in our current understanding of baseline testosterone and testosterone reactivity.

1.1.2.1. Baseline Testosterone

Lower baseline testosterone levels are hypothesized to be important for parenting. Lower baseline testosterone may enhance the likelihood that a parent engages in supportive and nurturing caregiving behaviors while reducing caregiving behaviors that are misaligned with the social demands of childrearing (e.g., aggression, risk taking, impulsivity, competition). For instance, lower baseline testosterone levels are associated with paternal sensitivity, nurturance, and cooperative behavior in males (Gettler et al. 2020; Weisman et al. 2014). Similarly, in a recent study of mothers and nonmothers, low levels of baseline testosterone were associated with longer viewing times of smiling and crying infants. This work suggests that in women, lower levels of baseline testosterone are associated with a higher approach motivation to soothe a distressed baby (Sinisalo et al. 2025). Furthermore, women with young children (under 3 years old) have lower levels of baseline testosterone than childless women and women with older children (Barrett et al. 2013). In addition, lower baseline testosterone levels are more common in parents and expecting fathers, relative to single nonparent men and women (Edelstein et al. 2017; Gray et al. 2006; Kuzawa et al. 2010), especially parents of young children (Barrett et al. 2013).

Baseline levels of testosterone may also be important for relationship outcomes. Specifically, men and women who are married or in committed relationships have lower baseline testosterone levels than individuals who are single (Gray et al. 2002; Holmboe et al. 2017; van Anders and Watson 2007). In a recent study of couples expecting their first child, a greater drop in fathers’ baseline testosterone levels from prenatal to postpartum was associated with higher relationship quality for both parents (Cárdenas et al. 2023). Similarly, fathers who had lower baseline testosterone during their partner's pregnancy reported higher relationship investment (Saxbe et al. 2017). However, other studies have found no association of baseline testosterone with marital relationship quality unless fathers experienced distress due to their role responsibilities (role overload; Booth et al. 2005). Interestingly, less is known about testosterone and relationship outcomes reported by mothers despite the calls from researchers to examine gender/sex differences in testosterone (Booth et al. 2006).

In sum, lower baseline testosterone levels are linked to parenting, but context may be important in understanding these associations. Furthermore, the association between baseline testosterone and specific relationship quality factors, such as social support, is less well understood.

1.1.2.2. Testosterone Reactivity

In addition to baseline levels of testosterone, testosterone reactivity—or changes in testosterone in response to a task or stressor—and its association with relationship outcomes during the transition to parenthood is not well understood. Single‐timepoint assessments (like baseline testosterone) are limited in their capacity to capture fluctuations in biomarker activity during periods of stress or conflict, which are particularly prevalent during the transition to parenthood, as physical routines (e.g., sleep) and relational needs (e.g., partner seeking help feeding the infant while cleaning garments) change constantly. In particular, testosterone changes that occur in response to parenting‐relevant tasks have the potential to tap into how a parent biologically responds during this relational transition.

To our knowledge, only a few studies have examined testosterone reactivity in couples in the context of a parenting‐specific stress task. In a study employing a stressor task, van Anders et al. (2012) examined how fathers respond to different conditions involving an interactive doll. Men experienced significant decreases in testosterone when they were able to interact with and successfully soothe the crying doll, whereas men who were able to hear the crying but were not able to interact or soothe the doll did not experience decreases in testosterone. In another study using the same interactive doll paradigm, Voorthuis et al. (2019) found that participants, all of whom were female, showed a decrease in testosterone while completing a 30‐min caregiving task with the doll, even when the caregiving task involved conflicting demands (e.g., caring for the baby while completing a lab task). In a study examining hormonal responses to an infant stimulator task, participants who used more affectionate touch with the infant doll had lower testosterone reactivity (Sinisalo et al. 2022). In support of the Steroid/Peptide Theory of Social Bonds (van Anders et al. 2011), these three studies provide evidence for the importance of context when examining testosterone and parenting.

1.2. The Importance of Examining Testosterone With Relationship Outcomes

The Challenge Hypothesis (Wingfield et al. 1990), first developed as a theory for understanding aggression within birds, has also been applied to human research (Wingfield 2017). According to this perspective, higher levels of testosterone are associated with aggression and dominance when individuals are “challenged” or in competitive situations. A recent meta‐analysis (Geniole et al., 2020) found that higher levels of baseline and change testosterone (here referred to as testosterone reactivity) were associated with higher aggression, especially in men. In the context of mating, higher levels of testosterone allow individuals to mate or protect their kin (Roney & Gettler, 2015). According to the theory, in situations that are nonthreatening or noncompetitive, testosterone levels should decrease. Thus, in the context of parenting, when mating is complete, testosterone levels should also decrease to allow individuals to support and care for their families (Wingfield 2017; Wingfield et al. 1990). However, when parenting elicits stress that can be perceived as a sense of “challenge” or competition, this could yield higher levels of testosterone. On the contrary, other parents may experience parenting stress as a signal to activate nurturance and responsiveness to the parenting situation. In other words, some individuals may respond to challenging parenting situations in a way that allows them to be supportive, while others may be more protective and dominant, creating variability in the degree of support that is provided and thus perceived between partners. We find this to be a logical deduction of previous findings that is supported by the mixed findings from Sinisalo et al. (2025), who examined approach and avoidance in response to infant distress. They argue that lower levels of testosterone may be related to positive parenting behavior (supporting the infant), while higher levels may be related to avoidance, or could also reflect protective parenting behaviors. Similarly, a meta‐analysis by Geniole et al. (2020) found that higher baseline testosterone and higher testosterone reactivity was associated with higher aggression, which was stronger in men than women. Examining testosterone at baseline and reactivity in association with perceptions of social support in couples across the transition to parenthood will allow for a nuanced understanding of the couple relationship.

1.3. The Current Study

In the current study, we examined the association of prenatal baseline testosterone and testosterone reactivity with self‐reported and partner‐reported postpartum perceived social support among first‐time parents. In so doing, we provide what we believe to be the first report of the links between testosterone and psychological perceptions, as prior research has only examined testosterone in relation to parenting behaviors (Edelstein et al. 2017; Endendijk et al. 2016; Gordon et al. 2017). Psychological perceptions provide important information regarding how behaviors are perceived within the couple.

Using rich, secondary dyadic data, we examined how each partner's (1) prenatal testosterone levels (baseline/basal) and (2) testosterone reactivity levels were associated with their own as well as their partner's postpartum perception of social support from each other. Testosterone levels were measured in expectant mothers (who were 6 months pregnant) and fathers before and after they both participated in a lab‐based co‐parenting task. Participants’ prenatal testosterone levels were measured at baseline/basal (or pre‐task), and participants’ testosterone reactivity levels were measured as the difference between the baseline/basal level and the post‐task level that was taken immediately following the parenting task.

1.3.1. Hypotheses

We tested the following two hypotheses, which were not preregistered. We hypothesized that lower prenatal baseline testosterone (Hypothesis 1) and lower prenatal testosterone reactivity (Hypothesis 2) among fathers would be associated with higher self‐reported and partner‐reported postpartum perceived social support. We also conducted exploratory analyses, testing whether prenatal baseline testosterone and prenatal testosterone reactivity among mothers predicted self‐reported and partner‐reported postpartum perceived social support. Given the limited prior work among women, we did not have directional hypotheses.

2. Method

2.1. Participants

Participants were cohabiting heterosexual couples (N couples = 110; Mothers: M age = 30.82, Fathers: M age = 32.41) who were 6 months pregnant, on average, with their first child (M gestation = 29.42 weeks, SD gestation = 2.69, rangegestation = 22–37 weeks). Couples were all racially White men and women. Parents were in good health without any neurological or psychological disorders and reported no substantial medication use or substance abuse. During the prenatal visit, the original study incurred loss of testosterone data due to technical issues, lack of saliva sampling kits, and time constraints. Please see Table 1 for final sample sizes for each variable. During the postpartum visit, mothers were about 6 months postpartum (infant age: M = 27.98 weeks, SD = 3.68, range = 21.98–42.76).

TABLE 1.

Correlation table.

N M (SD) Range 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1. Infant gender 101 — — 1
2. Mother education (years) 99 15.93 (2.90) 3.00–22.00 −0.02 1
3. Father education (years) 98 15.12 (2.98) 2.00–21.00 −0.09 0.27** 1
4. Mom age 105 30.82 (3.63) 23.31–42.76 −0.09 0.29** 0.01 1
5. Father age 105 32.41 (4.01) 23.34–42.21 −0.13 0.31** 0.20* 0.75*** 1
6. Prenatal visit time (minutes since midnight) 97 789.90 (221.70) 480–1200 0.07 0.03 −0.002 0.12 0.09 1
7. Mother smoker status 99 — — −0.09 0.30** 0.13 0.06 0.11 −0.22* 1
8. Mother weeks pregnant 107 29.42 (2.69) 22.00–37.00 −0.03 0.14 0.04 −0.10 −0.03 0.04 0.17 1
9. Mother fertility treatment 99 — — −0.03 0.11 0.15 0.24* 0.22* 0.10 0.09 0.08 1
10. Father smoker status 98 — — −0.07 0.03 0.19 −0.12 −0.09 0.06 0.24* 0.07 0.08 1
11. Mother prenatal baseline testosterone level 92 68.30 (22.77) 24.90–136.80 −0.10 0.18 0.18 −0.04 0.10 −0.02 0.19 0.31** 0.05 0.07 1
12. Father prenatal baseline testosterone level 91 97.61 (28.53) 44.20–191.10 −0.02 0.16 −0.04 0.07 0.09 −0.21 0.08 0.06 −0.002 −0.08 0.04 1
13. Mother prenatal post‐task testosterone level 87 79.33 (26.13) 23.00–139.80 −0.08 0.09 0.13 −0.12 0.04 −0.16 0.07 0.20 0.04 −0.06 0.69*** −0.04 1
14. Father prenatal post‐task testosterone level 90 107.83 (31.30) 42.70–194.60 −0.15 0.11 0.07 −0.05 0.03 −0.40*** 0.10 −0.01 0.05 −0.18 0.17 0.71*** 0.11 1
15. Mother prenatal testosterone reactivity level 87 11.83 (19.33) −41.10 to 71.50 0.05 −0.08 −0.02 −0.13 −0.10 −0.13 −0.12 −0.19 −0.02 −0.11 −0.23* −0.11 0.54*** −0.07 1
16. Father prenatal testosterone reactivity level 88 10.73 (22.99) −97.90 to 94.50 −0.15 −0.04 0.18 −0.16 −0.07 −0.28* 0.05 −0.09 0.14 −0.12 0.20 −0.29** 0.20 0.47*** 0.04 1
17. Mother prenatal perceived social support 107 6.69 (0.51) 4.50–7.00 −0.01 −0.18 −0.11 −0.08 −0.10 0.01 −0.09 0.06 0.06 0.02 −0.12 0.01 −0.09 −0.03 −0.01 −0.05 1
18. Father prenatal perceived social support 107 6.38 (0.71) 4.25–7.00 0.14 −0.04 0.06 −0.30** −0.28** 0.02 −0.12 0.03 −0.11 0.08 0.02 0.05 0.02 −0.08 0.05 −0.14 0.004 1
19. Mother postpartum perceived social support 100 6.40 (0.83) 3.25–7.00 0.08 −0.09 −0.09 −0.06 −0.13 0.01 −0.03 0.12 −0.07 0.09 −0.30** 0.15 −0.19 −0.12 0.05 −0.37*** 0.51*** 0.12 1
20. Father postpartum perceived social support 95 4.31 (2.45) 1.00–7.00 −0.02 −0.05 −0.07 −0.12 −0.12 0.18 −0.07 −0.21* −0.03 0.03 −0.08 −0.18 −0.16 −0.10 −0.05 0.08 −0.20 0.08 −0.17 1
Open in a new tab

Note. *p < 0.05, **p < 0.01, ***p < .001.

2.2. Procedure

Couples were invited to participate in a study on the transition to parenthood via internet groups and communities and through medical facilities. Couples who were interested in participating were invited to participate in a lab‐based assessment. This initial assessment was scheduled to occur 3 months prior to their expected delivery date. Couples provided consent, and the first saliva sample (indexing baseline [pre‐task] testosterone levels) was collected 5 min prior to a 10‐min standardized parenting stressor task, the inconsolable doll task (Shai 2018). Then, parents were asked to complete questionnaires for about 20 min separately before the second saliva sample. They were reunited after the second saliva sample was completed.

Couples were contacted again when the infant was 6 months old to participate in the postpartum follow‐up assessment, which included a battery of follow‐up questionnaires. All study procedures and measures were approved by the Institutional Review Board.

2.2.1. Inconsolable Doll Task

The inconsolable doll task was designed to be an ecologically valid representation of parenting stress and co‐parenting behaviors; specifically, it was designed to elicit negative emotion in participants relative to their baseline emotions (Shai 2018; Zamir et al. 2024). A research assistant joined the participants in the laboratory and introduced a doll to them. The doll looked and sounded like a real baby. The research assistant told the parents that the doll responded to caretaking just like a real baby would. Once the baby started to cry, the research assistant then demonstrated to the participants that they could soothe the baby by figuring out why it was crying (e.g., providing it with a bottle would soothe the baby if the baby was crying from hunger). Unbeknownst to the parents, after the research assistant left, the doll was programmed to be inconsolable, which was done to evoke parenting stress as the parents attempted to figure out what was bothering the doll (see Shai [2018] for a full description of the task).

The task involved the parents interacting with the doll, which was programmed to cry for 10 min, unbeknownst to participants. Parents were instructed to try to soothe the doll as they normally would. The use of this task created a realistic, high‐arousal context for the parent. For the first 5 min, one parent (randomly selected; referred to as the “task parent”) was left alone with the distressed “infant” and was instructed to soothe the baby, while the other parent (randomly selected; referred to as the “support parent”) waited outside. After 5 min had elapsed, the support parent then reunited with the task parent and was instructed to assist in soothing the child for the remaining 5 min. After the task, participants completed a battery of questionnaires. The second saliva sample was taken 20 min after the task ended (indexing post‐task testosterone levels).

2.3. Measures

2.3.1. Determination of Salivary Testosterone (Predictor)

Approximately 1 mL of whole saliva was collected by passive drool. All samples were refrigerated immediately, then stored at −80°C until assayed. Samples were assayed for testosterone in duplicate using a commercially available enzyme immunoassay kit (Salimetrics, Carlsbad, CA). The assay had a sample test volume of 25 µL, a range of sensitivity from 1 to 600 pg/mL, and average inter‐ and intraassay CVs below 15% and 10%, respectively.

Baseline testosterone was calculated as a single measure of salivary testosterone taken prior to the parenting stress task. Testosterone reactivity was calculated as a difference score, in which baseline testosterone levels were subtracted from post‐task testosterone levels following the parenting stressor (see Table 1 for descriptive statistics). A lower testosterone reactivity score represented less change from pre‐task (baseline) to post‐task. Importantly, a low reactivity score could have also resulted from a parent who had baseline testosterone levels that were either consistently low or high. We operationalize our measure of testosterone reactivity as the change in testosterone from pre‐task to post‐task (i.e., post‐task score − pre‐task score = change score).

2.3.2. Perceived Social Support (Outcome)

The four‐item “significant other” subscale from the Multidimensional Scale of Perceived Social Support (Zimet et al. 1988) was used to assess perceived adequacy of support given by a significant other (e.g., “I have a special person who is a real source of comfort to me.”). Both partners completed the subscale separately. Items were ranged from (1) Strongly Disagree to (7) Strongly Agree and were averaged to form a mean composite score of perceived social support. Higher scores indicated higher perceived social support from partner. Reliability was high at prenatal (α mother‐reported = 0.88; α father‐reported = 0.88) and postpartum (α mother‐reported = 0.91; α father‐reported = 0.99) assessments.

2.3.3. Demographics (Covariates)

At the baseline visit, parents were asked their infants’ sex assigned at birth (male [1], female [2]); their highest level of education (in years); their age (in years); and their salary (below average [1], average [2], and above average [3]).

2.3.4. Medical/Lifestyle Factors (Covariates)

We also measured several medical/lifestyle factors that may be correlated with salivary hormones: prenatal visit time (number of minutes since midnight), smoker status (yes [1], no [0]), mother weeks pregnant, and mother fertility treatment (yes [1], no [0]).

2.4. Analytic Strategy

2.4.1. Preliminary Analyses

Data analyses were conducted using a “wide” data structure that included a row of data for each dyad. Within each row, each participant's respective data (mother, father) were listed as a column of data. We ran Pearson's r correlations between demographics (child sex, mother and father years of education, mother and father age), medical/lifestyle factors reported prenatally (time of day during the lab session, mother and father smoker status, mother weeks pregnant, mother use of fertility treatment), and key study variables (mother and father prenatal baseline testosterone, mother and father prenatal post‐task testosterone, mother and father prenatal testosterone reactivity, and prenatal and postpartum perceived social support).

Next, using a long data structure where each participant was assigned a row of data, we examined how our sample responded to the inconsolable doll task. During the inconsolable doll task, testosterone may have increased or decreased from the pre‐task (baseline) saliva measure for participants. Using a crosstab analysis, we reported the frequency of participants who showed an increase in testosterone levels and the frequency of participants who showed a decrease in testosterone levels from the pre‐task to the post‐task saliva measures. This helped us characterize our sample to determine how mothers and fathers changed in their testosterone between these two timepoints.

Furthermore, using the long data structure, we conducted independent means t‐tests to further understand our sample: first, to determine if there were differences in key study variables between mothers and fathers; second, to determine if there were differences in key study variables between the roles that the partner was assigned in the inconsolable doll task—task‐parent (the parent who consoled the doll on their own) or support‐parent (the parent who entered at the 5‐min mark to support the task‐parent).

We did not correct for multiple tests in any of the preliminary analyses.

2.4.2. Primary Analyses

Primary analyses were conducted with MPlus v.8.10, which uses all available values and full information maximum likelihood (FIML) estimation, a preferred approach for handling missing data (Enders and Bandalos 2001). Model fit was assessed using comparative fit index (CFI) and root mean square error of approximation (RMSEA). Primary analyses were tested using the actor–partner interdependence model in a structural equation framework. In APIM analyses, the dyad is treated as the unit of analysis to separately model actor effects and partner effects within dyads. We distinguished the dyads by sexr, as all couples in our sample were  female and male soon‐to‐be parents. Actor effects referred to one individual's prenatal baseline testosterone levels and prenatal testosterone reactivity levels as a predictor of his/her own perception of postpartum perceived social support. Partner effects referred to the same individual's prenatal baseline testosterone levels and prenatal testosterone reactivity levels as a predictor of their partner's postpartum perceived social support.

In the APIM model, we included prenatal baseline testosterone levels and prenatal testosterone reactivity as predictor variables in the same model predicting postpartum perceived social support. Thus, an actor effect, for example, pertained to mothers’ baseline testosterone and testosterone reactivity as they related to mothers’ postpartum perceived social support and fathers’ baseline testosterone and testosterone reactivity as they related to fathers’ postpartum perceived social support. Similarly, a partner effect, for example, pertained to mothers’ predictor variable as it related to fathers’ outcome variable and fathers’ predictor variable as it related to mothers’ outcome variable.

Both partners’ baseline testosterone levels and testosterone reactivity levels were correlated with residuals of both partners’ report of perceived social support to account for the interdependence between dyad members.

In the APIM model (see Figure 1), we adjusted for prenatal perceived social support. This allowed us to predict postpartum perceived social support over and above how much support was perceived prenatally.

FIGURE 1.

FIGURE 1

Open in a new tab

Actor–partner interdependence model for baseline (Panel A) and reactivity (Panel B) testosterone predicting postpartum perceived social support after adjusting for prenatal perceived social support. For ease of interpretation, effects of covariates and residuals for the error terms are not shown. Correlations between exogenous variables are also not shown. *p < 0.05, **p < 0.01, and ***p < 0.001.

3. Results

3.1. Preliminary Analyses

3.1.1. Descriptive Statistics

Participants had average levels of baseline testosterone (M mother = 68.30, SD = 22.77; M father = 97.61, SD = 28.53), which, on average, increased post‐task (M mother = 79.33, SD = 26.13; M father = 107.83, SD = 31.30). Furthermore, mother and father pre‐task, mother and father post‐task, and mother testosterone reactivity levels were all normally distributed with skewness and kurtosis scores between 1 and −1. However, father testosterone reactivity was highly kurtotic (6.65) due to outliers. Therefore, we Winsorized outliers to 3 SD above and below the mean and presented all findings below using the Winsorized variable.

3.1.2. Correlations

Descriptive statistics and correlations between demographics, medical/lifestyle factors, and primary study variables are presented in Table 1. Correlations revealed that mother age was correlated with father prenatal perceived social support (r = −0.30, p = 0.002); time of day was correlated with father prenatal baseline testosterone (r = −0.21, p = 0.050), prenatal post‐task testosterone (r = −0.40, p < 0.001), and prenatal testosterone reactivity (r = −0.28, p = 0.010); number of weeks pregnant was correlated with mother prenatal baseline testosterone (r = 0.31, p < 0.003) and father postpartum perceived social support (r = −0.21, p = 0.038); and fertility treatment was correlated with mother age (r = 0.24, p = 0.016) and father age (r = 0.22, p = 0.033).

3.1.3. Test of Mean Differences

Next, we examined the frequency of participants who increased or decreased in testosterone levels from their pre‐task to post‐task samples and the significance of their decrease in perceived social support from the prenatal to postpartum timepoint. Most mothers and fathers demonstrated an increase in testosterone in response to the parenting task (n mothers = 66 [75.90%], n fathers = 67 [76.13%]). Furthermore, using a paired sample t‐test, both mothers (t(99) = −3.99, p < 0.001; Cohen's d = 0.72) and fathers (t(94) = −8.34, p < 0.001; Cohen's d = 2.48) reported significantly lower perceived social support at the postpartum timepoint relative to the prenatal timepoint. Table 2.

TABLE 2.

Actor–partner interdependence model.

Dependent variable
Father postpartum perceived social support Mother postpartum perceived social support
b SE b SE
Father testosterone reactivity 0.007 0.012 −0.010** 0.003
Mother testosterone reactivity −0.019 0.014 0.002 0.004
Father baseline testosterone −0.014 0.009 0.002 0.002
Mother baseline testosterone −0.013 0.012 −0.006* 0.003
Prenatal perceived social support 0.384 0.374 0.738*** 0.131
Open in a new tab

*p < 0.05, **p < 0.01, and ***p < 0.001.

Results of independent means t‐tests revealed that mothers (M = 68.30) had significantly lower prenatal baseline testosterone levels than fathers (M = 97.61) (t(181) = 7.68, p < 0.001; Cohen's d = 25.79). Mothers (M = 79.33) also had significantly lower prenatal post‐task testosterone levels than fathers (M = 107.83) (t(175) = 6.57, p < 0.001; Cohen's d = 28.87). However, testosterone reactivity scores (the difference between pre‐ and post‐task levels) did not significantly differ between parents (t(173) = −0.34, p = 0.733, Cohen's d = 21.25), and whether the participant showed an increase or decrease between timepoints also did not significantly differ between parents (t(173) = 0.04, p = 0.966, Cohen's d = 0.43). Parents did significantly differ on prenatal perceived social support (M mother = 6.69, M father = 6.38, t(214) = −3.76, p < 0.001, Cohen's d = .62) and postpartum perceived social support (M mother = 6.40, M father = 4.31, t(193) = −8.08, p < 0.001, Cohen's d = 1.81). Parents also significantly differed in how much they decreased in perceived social support from the prenatal to postpartum timepoint (M mother = 0.29, M father = 2.12, t(193) = 7.09, p < 0.001, Cohen's d = 1.80).

There were no significant differences between partner role (task‐parent vs. support‐parent) in prenatal baseline testosterone (p = 0.38), prenatal post‐task testosterone (p = 0.26), prenatal testosterone reactivity (p = 0.64p), whether participants increased or decreased between the timepoints (p = 0.27), prenatal perceived social support (p = 0.96), or postpartum perceived social support (p = 0.52). Furthermore, inclusion and exclusion of partner role in the models did not change the pattern of results. Therefore, partner role was not statistically adjusted for in the primary analyses.

3.1.4. Missingness Correlation

We examined whether missingness on key study variables was correlated with demographic variables. Missingness on mothers’ report of postpartum perceived social support was correlated with father age (r = 0.22, p = 0.026) and mother age (r = 0.22, p = 0.026) such that those who were missing data on postpartum perceived social support were older. Importantly, in our primary analyses, the FIML estimation was used, which runs analyses using available data and has been found to be the most superior method for handling missing data (Enders and Bandalos 2001).

3.1.5. Statistical Covariates

As a result of preliminary analyses, we identified mother and father age, time of day, and number of weeks pregnant as variables to be corrected for in the model.

3.2. Primary Analyses—Actor–Partner Interdependence Model

After including mother and father age, time of day, and number of weeks pregnant in the model (see Table 2), overall fit indices indicated that the model had acceptable fit, RMSEA = 0.000 (90% CI: 0.000, 0.049), CFI = 1.00.

3.2.1. Predicting Mother Postpartum Perceived Social Support

3.2.1.1. Actor Effects

The first actor effect was significant, such that after adjusting for prenatal perceived social support, mother prenatal baseline testosterone was associated with mother postpartum perceived social support (est = −0.006, SE = 0.003, p = 0.038). Mothers who had lower levels of prenatal baseline testosterone perceived more social support from their partners postpartum. The second actor effect of mother prenatal testosterone reactivity with mother postpartum perceived social support was not significant (est = 0.002, SE = .004, p = 0.599).

3.2.1.2. Partner Effects

The first partner effect of father prenatal baseline testosterone with mother postpartum perceived social support was not significant (est = 0.002, SE = .002, p = 0.397).

The second partner effect was significant such that after adjusting for prenatal perceived social support, father prenatal testosterone reactivity was associated with mother postpartum perceived social support (est = −0.01, SE = 0.003, p = 0.001). Fathers who had lower prenatal testosterone reactivity in response to the inconsolable doll task had partners who felt more supported postpartum.

3.2.2. Predicting Father Postpartum Perceived Social Support

There were no significant actor or partner effects of mother or father prenatal testosterone predicting father postpartum perceived social support (ps > 0.05).

4. Discussion

The current study followed first‐time parents across their transition to parenthood and measured their prenatal baseline testosterone and prenatal testosterone reactivity in response to a stressful simulated caregiving situation. We examined links between testosterone (both baseline levels and reactivity) and postpartum perceived social support. There were three noteworthy findings: (1) both mothers and fathers experienced significant declines in perceived social support between the prenatal and postpartum timepoints, (2) lower mother prenatal baseline testosterone was associated with higher mother postpartum perceived social support, and (3) lower father testosterone reactivity was associated with higher mother postpartum perceived social support. Of note, the only notable association of testosterone was its link with mothers’ perceptions of social support during the postpartum period. These findings suggest that both partners may contribute to changes in mothers’ perceptions of support during the transition to parenthood. In contrast, fathers’ perception of social support may be less influenced by biological changes in response to parenting stressors from either themselves or their partner.

4.1. Mothers’ Perception of Social Support

We found that mothers reported significantly lower perceived social support at the postpartum timepoint relative to what they reported prenatally. This finding adds to the mixed literature on social support changes across the transition to parenthood. One study found that while mothers’ reports of the quantity of support did not change from the prenatal to the postpartum period, perceived satisfaction with social support significantly decreased from prenatal to postpartum (L. H. Goldstein et al. 1996). A more recent study found no changes in perceived social support across the transition to parenthood (Saeieh et al. 2017). Overall, the mixed findings regarding changes to perceived support in mothers during this transition to parenthood might suggest individual differences or other factors that may buffer reductions in perceived social support. Understanding factors that can promote perceived social support is critical, which the current study begins to do.

Mothers’ own prenatal baseline testosterone was associated with their own postpartum perceived social support, after accounting for prenatal perceived social support. This finding is consistent with previous studies that found that lower levels of testosterone were associated with higher self‐ratings of satisfaction and commitment (Edelstein et al. 2017). Lower trait levels of testosterone may be adaptive for women when navigating potential relational and parenting stressors during the transition to parenthood. Although women generally have elevated levels of testosterone during pregnancy (Edelstein et al. 2017), those with lower levels may perceive more positive relational outcomes, perhaps because they can uphold a state of equanimity during what can be an emotionally turbulent time for first‐time parents. Lower testosterone may also suggest that mothers feel less threatened or challenged, allowing them to see their partner in a more positive light. Lower levels of testosterone may allow mothers to exhibit fewer dominant behaviors typical of elevated testosterone in mammals (Wingfield 2017). As such, male partners may feel less threatened and be more responsive to their needs, resulting in subsequent higher perceptions of support postnatally. Future work should examine how specific partner behaviors and responses, such as coping behaviors (Falconier et al. 2023), mediate the associations between testosterone levels and perceived social support.

Father's prenatal testosterone reactivity was associated with their partner's postpartum perceived social support, after adjusting for prenatal perceived social support. Those who have lower testosterone reactivity when responding to a parenting task may be able to engage in more supportive behavior, either in the realm of co‐parenting support or in terms of providing direct support to their partner (e.g., encouragement, reassurance) because they are feeling less frustrated and less physiologically activated. This argument is supported by more than 60 years of behavioral endocrinology research that has linked individual differences in testosterone to social behavior (e.g., Booth et al. 2006).

Contrary to public perception, it is widely believed among the scientific community that testosterone does not “cause or create” behavior; rather, testosterone is associated with social behavior via context contingency (Knight et al. 2022). That is, testosterone's presence, along with environmentally triggered changes in its levels, influences the probability that behavior patterns that already exist will be expressed given certain contextual demands and cues (Booth et al. 2000). In fact, Booth and colleagues argue that the social context of the family could be a key moderator of testosterone–behavior associations. Thus, in the current study, mothers may be able to view their male partners as more stable and responsive to their needs when their partner shows lower levels of testosterone reactivity in response to parenting stress. Alternatively, fathers who respond to a parenting challenge with an increase in testosterone may interpret parenting as a challenge. Such fathers may not provide the type of support that mothers need during the transition to parenthood. Importantly, although 21 fathers experienced a decrease in testosterone from pre‐task to post‐task, what we found is that less change in father testosterone from pre‐ to post‐task was associated with higher mother perceived social support. Thus, it seems that consistency in father testosterone levels (less change) is what was helpful for mothers. From an evolutionary perspective, being sensitive to indicators of partner support could confer an advantage for the child—a mother's attunement to the degree of support that a partner can provide may allow her to know whether her partner will be available to support her child for the long haul.

Our model did not reveal any actor effects of mother testosterone reactivity on mother postpartum perceived social support. Future studies may examine whether other indices of hormonal reactivity to a parenting task, or testosterone reactivity to other parenting tasks, predict mothers’ perceptions of social support during the postpartum phase. If future findings are consistent with ours, it may suggest that a mother's biological change to stress (and behavioral responses to these changes) is not as important for their own perceived social support but that their partner may play a bigger role.

Furthermore, our model did not reveal any partner effects of fathers’ prenatal baseline testosterone predicting mothers’ postpartum perceived social support. Perceived social support may function differently from behavioral (e.g., conflict or hostility) measures, which have been found to be associated with baseline levels of testosterone (Gettler et al. 2020; Schneiderman et al. 2014). Perhaps mothers have habituated to their partners’ baseline levels of testosterone, eliminating the association between the two. This is supported by our null correlation between father baseline testosterone and mother perceived social support. However, power and sample size may also contribute to null effects, so we interpret this finding with caution.

4.2. Understanding Father's Perception of Social Support

Interestingly, there were no paths in the actor–partner interdependence model that were significantly associated with father's perceived social support. Since mothers are more likely to be the primary caregivers during this transition to parenthood, social support may be more salient and sensitive for mothers than fathers during this period. In our sample, mothers were the primary caregiver most of the time (47%), followed by equal distribution of labor (40%) and fathers being the primary caregiver (8%). This idea is supported in a recent study examining maternal and paternal depression across the transition to parenthood. Father responsiveness (measured through reflective functioning) was protective against maternal depression, but mother responsiveness was not protective against paternal depression (A. Goldstein et al. 2022). Furthermore, downregulating negative emotions in wives is associated with marital quality for both partners (Bloch et al. 2014). Thus, it may be that fathers’ perception of social support during the transition to parenthood is more robust and not as easily influenced by their partner's physiological changes, which is important given how many physiological changes happen for women across the transition to parenthood.

Other factors may be more influential in predicting fathers’ perception of social support during this time. It may be that other measures are better at capturing father's experiences, while others are better at capturing mother's experiences. For instance, if fathers feel useful, valued, and wanted, then they may also feel supported. It also may be that different types of social support are more valued or needed for fathers versus mothers. The null actor effect is consistent with existing literature suggesting that low levels of male testosterone do not relate to male's own self‐reported perceptions of support (e.g., Edelstein et al. 2017). These findings highlight the importance of mother experiences for both partners and that fathers may be impacted less by their partner.

4.3. Strengths and Limitations

The current study has several strengths that should be noted. This is one of the first studies to examine physiology as a predictor of perceived social support, contributing to our understanding of the ways in which biological, psychological, and social factors may work together to predict health outcomes (see Adler [2009] for how the biopsychosocial model is still relevant today; see Engel [1977] for the development of the biopsychosocial model). Using a longitudinal design and accounting for prenatal perceived social support enhances our confidence that our findings reflect changes in perceived social support across the transition to parenthood. Use of a highly standardized and ecologically valid parenting task in a laboratory setting (Shai 2018) enhances the generalizability of the findings. We also examined our models among both men and women during an important developmental phase. Finally, incorporating biological data and self‐report data enabled us to avoid issues of shared method variance.

Nonetheless, study limitations should also be noted. First, our sample was an ethnically homogenous sample, limiting our ability to generalize our findings to diverse racial/ethnic groups. These research questions should be examined in future samples that include an ethnically and economically diverse sample of individuals. Second, our analyses did not adjust for some variables that have been previously associated with testosterone levels (e.g., BMI; Osuna et al. 2006), which may have accounted for some of the unexplained variance in the current model. Correlation analyses were conducted to offer an exploratory perspective on the data rather than to test hypotheses. As such, we did not adjust for multiple comparisons for the correlational analyses. In contrast, the APIM was used to test hypotheses; by design, this model includes four paths, which could inflate the possibility of a Type I error. Further work should continue to unpack these associations by using a more comprehensive model that includes other relevant factors as well as by examining mechanisms of these associations (e.g., behaviors such as cooperation, nurturance) and preregistered hypotheses. Third, observing parenting behavior in the laboratory rather than the natural environment (with a human infant) reduces the external validity of the findings (while enhancing the internal validity). Fourth, we did not include behavioral measures in this study, limiting our ability to make conclusions about how changes in testosterone were associated with parenting behaviors.

4.4. Future Directions

As the first of its kind, this study generates interesting questions that pave the way for future work on this topic. For instance, future research should examine the association between prenatal testosterone and postnatal testosterone levels. Since lower levels of testosterone in fathers are associated with being a better provider (Gettler et al. 2020), it would be important to examine testosterone levels postnatally, as well as to explore associations between postnatal testosterone levels and child outcomes. It is also important to understand changes in testosterone over a longer postpartum period. We know from the literature that testosterone changes in response to pair‐bonding (Gray et al. 2006, 2002; Grebe et al. 2019), but we are less certain of what this means after pair‐bonding has occurred and when changes in the family system are introduced (such as across the transition to parenthood). Importantly, there are some inconsistent findings of testosterone and pair‐bonding among fathers (Meijer et al. 2019). Thus, our understanding of testosterone across the transition to parenthood should be further explored in future research. Further, future studies leveraging a lab‐based approach, like the inconsolable doll task (Shai 2018), should observe how parents react without instructions to support one another throughout the task. Perhaps support provided in the moment during these coparenting interactions has more predictive power than trait‐like perceptions of social support across the transition to parenthood. In fact, not having to ask for support is associated with better mental health outcomes among individuals who are coping with chronic illness (Yang et al. 2015), while having to ask for help can put undue burden (e.g., feeling embarrassed to ask) on help‐seekers (Bohns and Flynn 2010). Therefore, utilizing an experimental procedure that tasks both parents to console the doll could provide additional information into the coparenting relationship. Further, examining interactions between the parents and their actual baby during the postpartum period may also yield interesting insights.

5. Conclusion

These findings contribute to the literature on the role of psychobiological processes in a key developmental phase (the transition to parenthood) by providing evidence for the association between prenatal testosterone and postpartum perceptions of social support among first‐time parents. To date, this is the first study to use a dyadic longitudinal design to investigate prenatal testosterone levels and relationship outcomes during the transition to parenthood. More research is needed to further understand the role of physiological biomarkers, as they relate to psychological and behavioral processes within a family context.

Conflicts of Interest

In the interest of full disclosure, D.A.G. is the founder and chief scientific advisor at Salimetrics LLC and Salivabio LLC. These relationships are managed by the policies of the committees on conflict of interest at the Johns Hopkins University School of Medicine and the University of California, Irvine.

Acknowledgments

We would like to thank the couples who participated in our study during their transition to parenthood. We would also like to thank the research assistants who helped us with data collection. This research was supported by grants from the Israeli Science Foundation (No. 1888/14), and the FP7‐PEOPLE‐2012‐IEF—Marie‐Curie Action: Intra‐European Fellowships for Career Development (IEF) under grant #300805. Dr. Jennifer A. Somers was supported by NIMH Individual National Research Service Award (F32‐132254A).

Froidevaux, N. M. , Lai J., Simon S. G., et al. 2025. “The Dyadic Association of Testosterone With Perceived Social Support in Couples Across the Transition to Parenthood.” Developmental Psychobiology 67, no. 5: e70078. 10.1002/dev.70078

Contributor Information

Nicole M. Froidevaux, Email: nfroidev@uci.edu.

Jessica L. Borelli, Email: jessica.borelli@uci.edu.

Data Availability Statement

The data that support the findings of this study are available from the corresponding authors upon reasonable request.

References

  1. Adler, R. H. 2009. “Engel's Biopsychosocial Model Is Still Relevant Today.” Journal of Psychosomatic Research 67, no. 6: 607–611. 10.1016/j.jpsychores.2009.08.008. [DOI] [PubMed] [Google Scholar]
  2. Barrett, E. S. , Tran V., Thurston S., et al. 2013. “Marriage and Motherhood Are Associated With Lower Testosterone Concentrations in Women.” Hormones and Behavior 63, no. 1: 72–79. 10.1016/j.yhbeh.2012.10.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Barry, J. A. , Hardiman P. J., Siddiqui M., and Thomas M.. 2011. “Meta‐Analysis of Sex Difference in Testosterone Levels in Umbilical Cord Blood.” Journal of Obstetrics and Gynaecology 31, no. 8: 697–702. 10.3109/01443615.2011.614971. [DOI] [PubMed] [Google Scholar]
  4. Bar‐Shachar, Y. , Lopata S., and Bar‐Kalifa E.. 2023. “Relationship Satisfaction During COVID‐19: The Role of Partners' Perceived Support and Attachment.” Family Relations 72, no. 1: 105–121. 10.1111/fare.12767. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bloch, L. , Haase C. M., and Levenson R. W.. 2014. “Emotion Regulation Predicts Marital Satisfaction: More Than a Wives' Tale.” Emotion 14, no. 1: 130–144. 10.1037/a0034272. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bohns, V. K. , and Flynn F. J.. 2010. ““Why Didn't You Just Ask?” Underestimating the Discomfort of Help‐Seeking.” Journal of Experimental Social Psychology 46, no. 2: 402–409. 10.1016/j.jesp.2009.12.015. [DOI] [Google Scholar]
  7. Booth, A. , Carver K., and Granger D. A.. 2000. “Biosocial Perspectives on the family.” Journal of Marriage and Family 62, no. 4: 1018–1034. 10.1111/j.1741-3737.2000.01018.x. [DOI] [Google Scholar]
  8. Booth, A. , Granger D. A., Mazur A., and Kivlighan K. T.. 2006. “Testosterone and Social Behavior.” Social Forces 85, no. 1: 167–191. 10.1353/sof.2006.0116. [DOI] [Google Scholar]
  9. Booth, A. , Johnson D. R., and Granger D. A.. 2005. “Testosterone, Marital Quality, and Role Overload.” Journal of Marriage and Family 67, no. 2: 483–498. 10.1111/j.0022-2445.2005.00130.x. [DOI] [Google Scholar]
  10. Cárdenas, S. I. , Tse W., León G., et al. 2023. “Prenatal Testosterone Synchrony in First‐Time Parents Predicts Fathers' Postpartum Relationship Quality.” Hormones and Behavior 156: 105440. 10.1016/j.yhbeh.2023.105440. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Castle, H. , Slade P., Barranco‐Wadlow M., and Rogers M.. 2008. “Attitudes to Emotional Expression, Social Support and Postnatal Adjustment in New Parents.” Journal of Reproductive and Infant Psychology 26, no. 3: 180–194. 10.1080/02646830701691319. [DOI] [Google Scholar]
  12. Cohen, S. , and Wills T. A.. 1985. “Stress, Social Support, and the Buffering Hypothesis.” Psychological Bulletin 98, no. 2: 310–357. 10.1037/0033-2909.98.2.310. [DOI] [PubMed] [Google Scholar]
  13. Cutrona, C. E. 1986. “Objective Determinants of Perceived Social Support.” Journal of Personality and Social Psychology 50, no. 2: 349–355. 10.1037/0022-3514.50.2.349. [DOI] [PubMed] [Google Scholar]
  14. Dennis, C.‐L. , and Ross L.. 2006. “Women's Perceptions of Partner Support and Conflict in the Development of Postpartum Depressive Symptoms.” Journal of Advanced Nursing 56, no. 6: 588–599. 10.1111/j.1365-2648.2006.04059.x. [DOI] [PubMed] [Google Scholar]
  15. Edelstein, R. S. , Chopik W. J., Saxbe D. E., Wardecker B. M., Moors A. C., and LaBelle O. P.. 2017. “Prospective and Dyadic Associations Between Expectant Parents' Prenatal Hormone Changes and Postpartum Parenting Outcomes.” Developmental Psychobiology 59, no. 1: 77–90. 10.1002/dev.21469. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Endendijk, J. J. , Hallers‐Haalboom E. T., Groeneveld M. G., et al. 2016. “Diurnal Testosterone Variability Is Differentially Associated With Parenting Quality in Mothers and Fathers.” Hormones and Behavior 80: 68–75. 10.1016/j.yhbeh.2016.01.016. [DOI] [PubMed] [Google Scholar]
  17. Enders, C. K. , and Bandalos D. L.. 2001. “The Relative Performance of Full Information Maximum Likelihood Estimation for Missing Data in Structural Equation Models.” Structural Equation Modeling: A Multidisciplinary Journal 8, no. 3: 430–457. 10.1207/S15328007SEM0803_5. [DOI] [Google Scholar]
  18. Engel, G. L. 1977. “The Need for a New Medical Model: A Challenge for Biomedicine.” Science 196, no. 4286: 129–136. 10.1126/science.847460. [DOI] [PubMed] [Google Scholar]
  19. Falconier, M. K. , Wojda‐Burlij A. K., Conway C. A., and Kim J.. 2023. “The Role of Emotion Regulation in Couples' Stress Communication and Dyadic Coping Responses.” Stress and Health 39, no. 2: 309–322. 10.1002/smi.3186. [DOI] [PubMed] [Google Scholar]
  20. Geniole, S. , Bird B., McVittie J., Purcell R., Archer J., and Carré J.. 2020. “Is Testosterone Linked to Human Aggression? A Meta‐Analytic Examination of the Relationship Between Baseline, Dynamic, and Manipulated Testosterone on Human Aggression.” Hormones and Behavior 123: 104644. 10.1016/j.yhbeh.2019.104644. [DOI] [PubMed] [Google Scholar]
  21. Gettler, L. T. , Lew‐Levy S., Sarma M. S., Miegakanda V., and Boyette A. H.. 2020. “Sharing and Caring: Testosterone, Fathering, and Generosity Among BaYaka Foragers of the Congo Basin.” Scientific Reports 10, no. 1: 15422. 10.1038/s41598-020-70958-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Gillis, A. , Gabriel B., Galdiolo S., and Roskam I.. 2019. “Partner Support as a Protection Against Distress During the Transition to Parenthood.” Journal of Family Issues 40, no. 9: 1107–1125. 10.1177/0192513X19832933. [DOI] [Google Scholar]
  23. Goldstein, A. , Borelli J. L., and Shai D.. 2022. “In Her Shoes: Partner Reflective Functioning Promotes Family‐Level Resilience to Maternal Depression.” Development and Psychopathology 35, no. 2: 958–971. 10.1017/S0954579422000189. [DOI] [PubMed] [Google Scholar]
  24. Goldstein, L. H. , Diener M. L., and Mangelsdorf S. C.. 1996. “Maternal Characteristics and Social Support Across the Transition to Motherhood: Associations With Maternal Behavior.” Journal of Family Psychology 10, no. 1: 60–71. 10.1037/0893-3200.10.1.60. [DOI] [Google Scholar]
  25. Gordon, I. , Pratt M., Bergunde K., Zagoory‐Sharon O., and Feldman R.. 2017. “Testosterone, Oxytocin, and the Development of Human Parental Care.” Hormones and Behavior 93: 184–192. 10.1016/j.yhbeh.2017.05.016. [DOI] [PubMed] [Google Scholar]
  26. Gray, P. B. , Jeffrey Yang C.‐F., and Pope H. G.. 2006. “Fathers Have Lower Salivary Testosterone Levels Than Unmarried Men and Married Non‐Fathers in Beijing, China.” Proceedings of the Royal Society B: Biological Sciences 273, no. 1584: 333–339. 10.1098/rspb.2005.3311. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Gray, P. B. , Kahlenberg S. M., Barrett E. S., Lipson S. F., and Ellison P. T.. 2002. “Marriage and Fatherhood Are Associated With Lower Testosterone in Males.” Evolution and Human Behavior 23, no. 3: 193–201. 10.1016/S1090-5138(01)00101-5. [DOI] [Google Scholar]
  28. Grebe, N. M. , Sarafin R. E., Strenth C. R., and Zilioli S.. 2019. “Pair‐Bonding, Fatherhood, and the Role of Testosterone: A Meta‐Analytic Review.” Neuroscience and Biobehavioral Reviews 98: 221–233. 10.1016/j.neubiorev.2019.01.010. [DOI] [PubMed] [Google Scholar]
  29. Handelsman, D. J. , Sikaris K., and Ly L. P.. 2016. “Estimating Age‐Specific Trends in Circulating Testosterone and Sex Hormone‐Binding Globulin in Males and Females Across the Lifespan.” Annals of Clinical Biochemistry 53, no. 3: 377–384. 10.1177/0004563215610589. [DOI] [PubMed] [Google Scholar]
  30. Holmboe, S. A. , Priskorn L., Jørgensen N., et al. 2017. “Influence of Marital Status on Testosterone Levels—A Ten Year Follow‐Up of 1113 Men.” Psychoneuroendocrinology 80: 155–161. 10.1016/j.psyneuen.2017.03.010. [DOI] [PubMed] [Google Scholar]
  31. Kluwer, E. S. 2010. “From Partnership to Parenthood: A Review of Marital Change Across the Transition to Parenthood.” Journal of Family Theory & Review 2, no. 2: 105–125. 10.1111/j.1756-2589.2010.00045.x. [DOI] [Google Scholar]
  32. Knight, E. L. , Morales P. J., Christian C. B., et al. 2022. “The Causal Effect of Testosterone on Men's Competitive Behavior Is Moderated by Basal Cortisol and Cues to an Opponent's Status: Evidence for a Context‐Dependent Dual‐Hormone Hypothesis.” Journal of Personality and Social Psychology 123, no. 4: 693–716. 10.1037/pspa0000305. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Kuzawa, C. W. , Gettler L. T., Huang Y., and McDade T. W.. 2010. “Mothers Have Lower Testosterone Than Non‐Mothers: Evidence From the Philippines.” Hormones and Behavior 57, no. 4–5: 441–447. 10.1016/j.yhbeh.2010.01.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Lawrence, E. , Rothman A. D., Cobb R. J., Rothman M. T., and Bradbury T. N.. 2008. “Marital Satisfaction Across the Transition to Parenthood.” Journal of Family Psychology 22, no. 1: 41–50. 10.1037/0893-3200.22.1.41. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Maher, M. J. , Mora P. A., and Leventhal H.. 2006. “Depression as a Predictor of Perceived Social Support and Demand: A Componential Approach Using a Prospective Sample of Older Adults.” Emotion 6, no. 3: 450–458. 10.1037/1528-3542.6.3.450. [DOI] [PubMed] [Google Scholar]
  36. Maneshi Azghandi, A. , Pashang S., and Khodaverdian S.. 2022. “Predicting Perceived Social Support Based on Perceived Stress and Cognitive Emotion Regulation in Patients With Ulcerative Colitis.” Iranian Journal of Health Psychology 5, no. 1: 39–48. 10.30473/ijohp.2022.55667.1138. [DOI] [Google Scholar]
  37. Meijer, W. M. , van IJzendoorn M. H., and Bakermans‐Kranenburg M. J.. 2019. “Challenging the Challenge Hypothesis on Testosterone in Fathers: Limited Meta‐Analytic Support.” Psychoneuroendocrinology 110: 104435. 10.1016/j.psyneuen.2019.104435. [DOI] [PubMed] [Google Scholar]
  38. Mikulincer, M. , and Shaver P. R.. 2009. “An Attachment and Behavioral Systems Perspective on Social Support.” Journal of Social and Personal Relationships 26, no. 1: 7–19. 10.1177/0265407509105518. [DOI] [Google Scholar]
  39. Nawaz, S. , Javeed S., Haneef A., Tasaur B., and Khalid I.. 2014. “Perceived Social Support and Marital Satisfaction Among Love and Arranged Marriage Couples.” International Journal of Academic Research and Reflection 2, no. 2: 41–50. [Google Scholar]
  40. Osuna, J. , Gomez‐Perez R., Arata‐Bellabarba G., and Villaroel V.. 2006. “Relationship Between BMI, Total Testosterone, Sex Hormone‐Binding‐Globulin, Leptin, Insulin and Insulin Resistance in Obese Men.” Archives of Andrology 52, no. 5: 355–361. 10.1080/01485010600692017. [DOI] [PubMed] [Google Scholar]
  41. Pourmand, V. , Froidevaux N. M., Williams D. P., Yim I. S., and Campos B.. 2023. “Attachment Insecurity, Heart Rate Variability, and Perceived Social Support in a Diverse Sample of Young Adults.” Frontiers in Psychology 14: 1208924. 10.3389/fpsyg.2023.1208924. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Roney, J. R. , and Gettler L. T.. 2015. “The Role of Testosterone in Human Romantic Relationships.” Current Opinion in Psychology 1: 81–86. 10.1016/j.copsyc.2014.11.003. [DOI] [Google Scholar]
  43. Saeieh, S. E. , Rahimzadeh M., Yazdkhasti M., and Torkashvand S.. 2017. “Perceived Social Support and Maternal Competence in Primipara Women During Pregnancy and After Childbirth.” International Journal of Community Based Nursing and Midwifery 5, no. 4: 408–416. [PMC free article] [PubMed] [Google Scholar]
  44. Saxbe, D. E. , Edelstein R. S., Lyden H. M., Wardecker B. M., Chopik W. J., and Moors A. C.. 2017. “Fathers' Decline in Testosterone and Synchrony With Partner Testosterone During Pregnancy Predicts Greater Postpartum Relationship Investment.” Hormones and Behavior 90: 39–47. 10.1016/j.yhbeh.2016.07.005. [DOI] [PubMed] [Google Scholar]
  45. Schneiderman, I. , Kanat‐Maymon Y., Zagoory‐Sharon O., and Feldman R.. 2014. “Mutual Influences Between Partners' Hormones Shape Conflict Dialog and Relationship Duration at the Initiation of Romantic Love.” Social Neuroscience 9, no. 4: 337–351. 10.1080/17470919.2014.893925. [DOI] [PubMed] [Google Scholar]
  46. Shai, D. 2018. “The Inconsolable Doll Task: Prenatal Coparenting Behavioral Dynamics Under Stress Predicting Child Cognitive Development at 18 Months.” Infant Behavior and Development 56: 101254. 10.1016/j.infbeh.2018.04.003. [DOI] [PubMed] [Google Scholar]
  47. Sinisalo, H. , Bakermans‐Kranenburg M. J., and Peltola M. J.. 2022. “Hormonal and Behavioral Responses to an Infant Simulator in Women With and Without Children.” Developmental Psychobiology 64, no. 7: e22321. 10.1002/dev.22321. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Sinisalo, H. , Hahn A. C., Jones B. C., Bakermans‐Kranenburg M. J., and Peltola M. J.. 2025. “Impact of Parity and Salivary Hormonal Levels on Motivation Toward Infant Emotions.” Emotion 25, no. 4: 940–950. 10.1037/emo0001467. [DOI] [PubMed] [Google Scholar]
  49. Stanton, S. C. E. , and Campbell L.. 2014. “Psychological and Physiological Predictors of Health in Romantic Relationships: An Attachment Perspective: Attachment and Health.” Journal of Personality 82, no. 6: 528–538. 10.1111/jopy.12056. [DOI] [PubMed] [Google Scholar]
  50. Stapleton, L. R. T. , Schetter C. D., Westling E., et al. 2012. “Perceived Partner Support in Pregnancy Predicts Lower Maternal and Infant Distress.” Journal of Family Psychology 26, no. 3: 453–463. 10.1037/a0028332. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Swickert, R. J. , Hittner J. B., and Foster A.. 2010. “Big Five Traits Interact to Predict Perceived Social Support.” Personality and Individual Differences 48, no. 6: 736–741. 10.1016/j.paid.2010.01.018. [DOI] [Google Scholar]
  52. Ter Kuile, H. , Finkenauer C., Van der Lippe T., and Kluwer E. S.. 2021. “Changes in Relationship Commitment Across the Transition to Parenthood: Pre‐Pregnancy Happiness as a Protective Resource.” Frontiers in Psychology 12: 132. 10.3389/fpsyg.2021.622160. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Theiss, J. A. , Estlein R., and Weber K. M.. 2013. “A Longitudinal Assessment of Relationship Characteristics That Predict New Parents' Relationship Satisfaction.” Personal Relationships 20, no. 2: 216–235. 10.1111/j.1475-6811.2012.01406.x. [DOI] [Google Scholar]
  54. Tietjen, A. M. 1985. “The Social Networks and Social Support of Married and Single Mothers in Sweden.” Journal of Marriage and the Family 47, no. 2: 489–496. 10.2307/352148. [DOI] [Google Scholar]
  55. Uchino, B. N. 2009. “Understanding the Links Between Social Support and Physical Health: A Life‐Span Perspective With Emphasis on the Separability of Perceived and Received Support.” Perspectives on Psychological Science 4, no. 3: 236–255. 10.1111/j.1745-6924.2009.01122.x. [DOI] [PubMed] [Google Scholar]
  56. van Anders, S. M. , Goldey K. L., and Kuo P. X.. 2011. “The Steroid/Peptide Theory of Social Bonds: Integrating Testosterone and Peptide Responses for Classifying Social Behavioral Contexts.” Psychoneuroendocrinology 36, no. 9: 1265–1275. 10.1016/j.psyneuen.2011.06.001. [DOI] [PubMed] [Google Scholar]
  57. van Anders, S. M. , Tolman R. M., and Volling B. L.. 2012. “Baby Cries and Nurturance Affect Testosterone in Men.” Hormones and Behavior 61, no. 1: 31–36. 10.1016/j.yhbeh.2011.09.012. [DOI] [PubMed] [Google Scholar]
  58. van Anders, S. M. , and Watson N. V.. 2007. “Testosterone Levels in Women and Men Who Are Single, in Long‐Distance Relationships, or Same‐City Relationships.” Hormones and Behavior 51, no. 2: 286–291. 10.1016/j.yhbeh.2006.11.005. [DOI] [PubMed] [Google Scholar]
  59. Voorthuis, A. , Bakermans‐Kranenburg M., and Van IJzendoorn M.. 2019. “Testosterone Reactivity to Infant Crying and Caregiving in Women: The Role of Oral Contraceptives and Basal Cortisol.” Infant Behavior and Development 56: 101191. 10.1016/j.infbeh.2017.08.002. [DOI] [PubMed] [Google Scholar]
  60. Weisman, O. , Zagoory‐Sharon O., and Feldman R.. 2014. “Oxytocin Administration, Salivary Testosterone, and Father–Infant Social Behavior.” Progress in Neuro‐Psychopharmacology and Biological Psychiatry 49: 47–52. 10.1016/j.pnpbp.2013.11.006. [DOI] [PubMed] [Google Scholar]
  61. Wingfield, J. C. 2017. “The Challenge Hypothesis: Where It Began and Relevance to Humans.” Hormones and Behavior 92: 9–12. 10.1016/j.yhbeh.2016.11.008. [DOI] [PubMed] [Google Scholar]
  62. Wingfield, J. C. , Hegner R. E., A. M. Dufty, Jr. , and Ball G. F.. 1990. “The “Challenge Hypothesis”: Theoretical Implications for Patterns of Testosterone Secretion, Mating Systems, and Breeding Strategies.” The American Naturalist 136, no. 6: 829–846. 10.1086/285134. [DOI] [Google Scholar]
  63. Yang, J. P. , Leu J., Simoni J. M., Chen W. T., Shiu C.‐S., and Zhao H.. 2015. ““Please Don't Make Me Ask for Help”: Implicit Social Support and Mental Health in Chinese Individuals Living With HIV.” AIDS and Behavior 19: 1501–1509. 10.1007/s10461-015-1041-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Zamir, O. , Oved N., Szepsenwol O., et al. 2024. “The Intersection Between Alexithymia, Testosterone Reactivity, and Coparenting in Fathers Predicts Child's Prosocial Behavior.” Hormones and Behavior 164: 105565. 10.1016/j.yhbeh.2024.105565. [DOI] [PubMed] [Google Scholar]
  65. Zimet, G. D. , Dahlem N. W., Zimet S. G., and Farley G. K.. 1988. “The Multidimensional Scale of Perceived Social Support.” Journal of Personality Assessment 52, no. 1: 30–41. 10.1207/s15327752jpa5201_2. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The data that support the findings of this study are available from the corresponding authors upon reasonable request.

Articles from Developmental Psychobiology are provided here courtesy of Wiley

RESOURCES