Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Jul 12.
Published in final edited form as: Biol Psychol. 2021 Jul 12;163:108148. doi: 10.1016/j.biopsycho.2021.108148

The rewards of motherhood: Neural response to reward in pregnancy prospectively predicts maternal bonding with the infant in the postpartum period

Elizabeth M Mulligan 1, Magen Lowe 1, Heather Flynn 2, Greg Hajcak 1,3
PMCID: PMC8328945  NIHMSID: NIHMS1723616  PMID: 34265368

Abstract

Mother-to-infant attachment is critical to the health of mothers and offspring. While reward circuitry is implicated in maternal attachment, no studies have yet examined whether antenatal (i.e., in pregnancy) reward responsiveness predicts mother-to-infant bonding in the postnatal period. In a sample of 63 women, we examined whether the Reward Positivity (RewP), an event-related potential elicited to feedback indicating monetary reward, measured in pregnancy prospectively predicts self-reported mother-to-infant attachment at approximately one year postpartum. An increased antenatal RewP was associated with increased postnatal pleasure in proximity with the infant. Furthermore, this association was independent of associations with depression, anxiety, and stress symptoms. This is the first study to find a prospective association between reward responsiveness in pregnancy and postnatal bonding with the infant. Future directions for attachment research are discussed.

Keywords: reward positivity, attachment, event-related potential, maternal behavior, pregnancy, postpartum

Introduction

A mother’s bond with her infant is both critical to the health and development of offspring and to maternal mental health and well-being. Mother-infant bonding refers to a process in which a mother shows an affectional attachment to the infant, and thereby protects and cares for the infant (Broad et al., 2006). As such, mother-to-infant bonding plays an important role in maternal sensitivity to the infant’s needs, and can impact later attachment quality (Lewis & Feiring, 1989). Disorganized attachment is associated with increased risk for poor developmental outcomes in early childhood and beyond, including cognitive and socioemotional developmental delays (Groh et al., 2017; Alhusen et al., 2013), poor peer relationships (Groh et al., 2014; Sroufe et al., 2005; Jacobvitz & Hazen, 1999), emotional and behavioral dysregulation (Sroufe, 2005), and several forms of childhood psychopathology (Groh et al., 2012; DeKlyen & Greenberg, 2008). Furthermore, a reduced mother-to-infant bond is linked to adverse psychological outcomes for mothers, including maternal depression (Lindgren, 2001; Atkinson et al., 2000). Considering these adverse and pervasive consequences, it is important to examine precursors to reduced mother-to-infant attachment to inform future efforts for determining at-risk individuals.

Enhanced mother-infant bonding is thought to underlie maternal sensitivity (i.e., a mother’s ability to respond to her infant’s emotions and communications; Broad et al., 2006; Ainsworth, 1979) and recent studies suggest that pregnancy is associated structural and functional changes to neural areas underlying attentional allocation to social cues and social information processing—changes that are thought to facilitate maternal sensitivity and bonding (Gholampour et al., 2020; Hoekzema et al., 2016; De Carli et al., 2019). For instance, a previous study investigating changes in event-related potentials (ERPs) from pregnancy to the postpartum period found that increases in the P1 and P2 ERP components to infant faces—components thought to reflect attentional allocation— predicted stronger postnatal mother-to-infant bonding (Dudek et al., 2020). Furthermore, women who showed greater attentional bias toward infant distress during late pregnancy reported more successful mother-infant relationships in the postnatal period (Pearson et al., 2011). Recent work suggests that these structural and functional neural changes associated with pregnancy include changes to reward circuitry (Hoekzema et al., 2020). As such, the perinatal period is characterized by neural changes that facilitate increased attention and motivated behavior toward the infant, and thus, increased sensitive maternal care.

Support for the notion that reward circuitry plays a crucial role in sensitive maternal behavior comes from both the animal and human literatures. Mothers’ responses to infants rely on motivational systems including dopamine- (DA) and oxytocin- (OT) rich pathways in the dopaminergic reward system (Numan, 2007; Strathearn et al., 2009; Rincon-Cortes & Grace, 2020). Previous work in rodent models suggests that OT—released during activities such as breast-feeding and seeing, hearing, or touching the infant—acts on the medial preoptic area (MPOA), which projects to the ventral tegmental area (VTA), an area containing DA neurons that gives rise to the mesolimbic DA system (Numan, 2007). Hence, the MPOA to VTA circuit may stimulate release of dopamine to prefrontal neural regions to activate proactive maternal behavior (Numan, 2007) and OT may sensitize the reward system to infant-related cues (Strathearn et al., 2009; Olazabal & Young, 2006).

Rodent models have provided robust evidence for the notion that the mesolimbic DA system is critical for maternal responsiveness. Experiments leveraging radiofrequency, electrolytic, or neurotoxin lesions in DA-rich neural structures that comprise the mesolimbic DA pathway, including the ventral tegmental area and nucleus accumbens, have shown that these lesions impair rodent maternal behavior, such as nursing (Smith & Holland, 1975), nest building (Gaffori & Le Moal, 1979), and pup retrieval (Gaffori & Le Moal, 1979; Numan & Smith, 1984; Hansen et al., 1991), and may cause aggressive behavior toward pups (Smith & Holland, 1975; Gaffori & Le Moal, 1979). Furthermore, injection of a DA antagonist in the VTA also resulted in deficits in maternal behavior (Hansen et al., 1991). Finally, a study utilizing in vivo voltammetry to monitor changes in extra-cellular DA in the nucleus accumbens found that DA increases preceded pup licking and grooming and the duration of the DA increase correlated with the duration of the licking and grooming bout (Champagne et al., 2004).

Studies in postnatal women have yielded parallel findings (see Swain et al., 2014 for a review). In an fMRI study of human mothers, DA-rich reward processing regions, including the VTA, substantia nigra, and ventral striatum, were activated selectively when viewing one’s own infant’s face as compared to an unknown infant’s face (Strathearn et al., 2008). In a separate fMRI study, the peripheral oxytocin response to infant contact was positively correlated with ventral striatum response to images of one’s own baby (Strathearn et al., 2009), aligning with research in animal models that suggests oxytocin plays a role in maternal behavior via its effects on the mesolimbic DA system. In keeping with these findings, human mothers with drug and alcohol addictions demonstrate reduced activation in the ventral striatum to images of their own baby, as compared to mothers with no substance use problems (Kim et al., 2017), suggesting that drug addiction may disrupt reward-related neural systems that typically facilitate dopaminergic responses to infant cues. Similarly, compared to non-depressed mothers, depressed mothers showed reduced activity in the reward circuit, including the caudate and nucleus accumbens, when exposed to their own infant’s cry versus a generic baby-cry (Laurent & Ablow, 2012).

Given the substantial evidence for associations between reward circuitry and maternal behavior, reward function may have effects on mother-to-infant bonding and attachment. Indeed, reduced neural reward responsivity is associated with differences in attachment quality in the postnatal period. In an fMRI study of human mothers, mothers with secure attachment to their infants were characterized by greater activation of the ventral striatum in response to images of their own baby crying and laughing, as compared to dismissing mothers (Strathearn et al., 2009). Furthermore, as compared to non-depressed mothers, a group of depressed mothers showed both faster attenuation of the ventral striatum response to monetary reward and lower scores on a self-report measure indexing pleasure in interaction with their infant (Moses-Kolko et al., 2011).

While recent studies have found cross-sectional associations between maternal behavior and attachment and reward-related neural systems, there is a surprising lack of research examining prospective associations, especially in the antenatal period (i.e., in pregnancy). Reduced reward-related neural activity measured in the antenatal period may serve as a risk factor for reduced maternal-infant bonding in the postnatal period. Reliance on fMRI for examining brain activity is a barrier to this type of research, as fMRI is inappropriate for studying pregnant women due to unknown risks for both women and the fetus.

On the other hand, event-related potentials (ERPs) measured via electroencephalography (EEG) are a tool for examining neural activity that pose minimal risk to pregnant women and their offspring. Approximately 300 milliseconds (ms) after feedback indicating monetary reward, the ERP at frontocentral EEG recording sites is characterized by a relative positivity in amplitude; an apparent negativity is observed following feedback indicating monetary loss (Hajcak, Moser, Holroyd, & Simons, 2007). The ERP response to reward is referred to as the Reward Positivity (RewP; Baker & Holroyd, 2011; Proudfit, 2015). The RewP is associated with self-report and behavioral measures of reward responsivity (Bress & Hajcak, 2013) and with fMRI measures of activity in the ventral striatum in response to monetary reward (Carlson et al., 2011). Additionally, the RewP has excellent test-retest reliability and internal consistency (Levinson et al., 2017). Thus, the RewP is an ideal measure for studying reward responsivity in pregnant women given its excellent psychometric properties and safety for use in pregnancy.

Additionally, while previous studies have examined activity of reward-related regions to infant cues and its relationships with maternal bonding and attachment, there is a dearth of research examining whether more generic reward-related neural activity (e.g., to monetary rewards) might also be related to maternal bonding. Given that monetary and social rewards both activate the striatum, a key reward-related neural region (Izuma et al., 2008; Lin et al., 2012), individual differences in reactivity to monetary rewards during pregnancy may serve as an indicator of general reward system responsivity which could potentially predict the degree to which the infant is rewarding, and thus, maternal-infant bonding, in the postnatal period.

The present study is the first to test whether reward-related neural activity in pregnancy prospectively predicts self-reported mother-to-infant attachment in the postnatal period. To this end, we examined longitudinal associations between the RewP collected in pregnancy and self-reported mother-to-infant bonding in the postnatal period. Maternal bonding with the infant was assessed via the Maternal Postnatal Attachment Scale (MPAS; Condon & Corkindale, 1998), which included subscales indexing tolerance and acceptance of the infant and pleasure in proximity with the infant. We hypothesized that a larger antenatal RewP response would be associated with a higher scores on both subscales of the MPAS. We also sought to examine whether this association would exist independent of associations with anxious and depressive symptoms.

Method

Participants

237 women participated in the baseline session during pregnancy. 63 women with a mean age of 30.94 years (SD = 5.09) participated in the follow-up session in the postnatal period and were provided monetary compensation for participation.1 Descriptive statistics summarizing demographics are presented in Table 1. Women were recruited from North Florida Women’s Care, a local obstetrics and gynecology (OB/GYN) clinic in Tallahassee, FL, using flyers distributed by the front desk staff. Information about the study was listed on the flyer along with space for participants to provide contact information if interested in participation. Research staff contacted interested patients with more information and to schedule study visits. For the baseline visit in pregnancy, the mean gestational age for the sample was 21.17 weeks (SD = 8.85). For the follow-up visit in the postpartum period, research staff contacted participants after reaching 9 months postpartum to assess interest in completing follow-up surveys via email. At follow-up, participants had a mean age of 32.54 years (SD = 5.07) and were at a mean of 59.87 weeks postpartum (SD = 16.93). All women gave consent before completing the study and all research procedures were approved by the Institutional Review Board at Florida State University.

Table 1.

Demographics (top) and MPAS Scale Scores (bottom)

M SD
Demographics
 Age (years) at baseline 30.94 5.09
 Gestational weeks at baseline 21.17 8.85
 Gestational weeks at birth 39.13 1.56
 Postpartum weeks at follow-up 59.88 16.93
 Number of children at follow-up 1.71 1.11
 Race
  Asian 1.6%
  Black 11.1%
  Caucasian 76.2%
  Latino 7.9%
  Mixed Race 1.6%
  Pacific Islander 1.6%
M SD
MPAS Pleasure in Proximity 35.18 3.31
MPAS Tolerance and Acceptance 20.21 2.94
Open in a new tab

Measures

Maternal perceptions of attachment quality were assessed only at the follow-up assessment using the Maternal Postnatal Attachment Scale (MPAS; Condon & Corkindale, 1998). The MPAS is a 19-item inventory assessing maternal attachment to one’s own infant. Each item was rated for how closely the respondent felt the statements in the inventory aligned with their own feelings about caring for their infant. This measure has been shown to have high test-retest reliability and internal consistency, and is shown to have good construct validity (Condon & Corkindale, 1998). The present study examined three subscales within the MPAS: the Tolerance and Acceptance scale, the Competence scale, and the Pleasure in Proximity scale. Scale totals were computed by totaling subscales. A participant’s score on any particular scale was dropped from analysis if more than 15% of the items in the given scale were skipped. Cronbach’s alpha in the present sample for the Tolerance and Acceptance subscale, the Pleasure in Proximity subscale, and the Competence subscale were .68, .67, and .24, respectively. Due to the low reliability of the Competence scale in the present sample, the Competence subscale score was excluded from analyses.

Additionally, symptoms of depression and anxiety were assessed at the baseline and follow-up assessments using the Depression Anxiety Stress Scales (DASS-21; Lovibond & Lovibond, 1995). The DASS-21 is a 21-item inventory assessing symptoms of depression, anxiety, tension, and stress, and is comprised of three scales—the depression, anxiety, and stress scales. This measure is shown to have acceptable to excellent internal consistency and validity (Antony et al., 1998). A participant’s score on any particular scale was dropped from analysis if more than 15% of the items in the given scale were skipped. Cronbach’s alpha in the present sample for the depression, anxiety, and stress scales at the baseline assessment were .76, .77, and .86, respectively. Cronbach’s alpha for the depression, anxiety, and stress scales at the follow-up assessment were .72, .73, and .82, respectively.

Procedure

These data are part of a larger, longitudinal, and ongoing study on neural and psychosocial correlates of perinatal psychological functioning. For their convenience, participants had the option to participate in the baseline visit either at the local OB/GYN clinic before or after a prenatal doctor visit, or at the Psychology Department at Florida State University. Written documentation of consent was first obtained, and then self-report measures were administered electronically. After self-report measures were complete, EEG setup and recording began. The doors task (described below) was the first task performed while recording EEG; two other tasks followed in a fixed order, and those data will be reported elsewhere. The follow-up assessment either took place remotely via online surveys or during a visit to the laboratory at Florida State University, depending on the participant’s preference.

Doors Task

To measure the neural activity associated with reward sensitivity during the baseline assessment, participants completed a simple guessing paradigm on the computer while EEG was recorded. Presentation software (Neurobehavioral Systems, Inc., Albany, Ca, USA) was used to administer a version of the doors task that is similar to those used previously (Proudfit, 2015). A previous study that has examined the internal consistency reliability of the win and loss responses in the same task (i.e., the Doors task) as a function of the number of trials in an adult sample found that acceptable reliability (i.e., Cronbach’s alpha > .70) was reached by 15 to 20 trials (Levinson et al., 2017). This version of the doors task consisted of two blocks of 20 trials and began with the presentation of two identical doors. Participants were instructed to choose out of the two doors which door they believe had a monetary prize behind it. The participants were told they could win $0.50 or lose $0.25 on each trial. Participants had equal probability of receiving either gain or loss feedback (20 each), with the monetary value of a gain being twice as large as the monetary value of a loss to equalize the subjective value of monetary gains and losses (Tversky & Kahneman, 1981; Tversky & Kahneman, 1992). Participants were instructed to try to earn as much money as possible and that they would receive their earnings from the task in cash. At the start of the task, participants were presented with the two identical doors which remained presented until a selection was made. Once a door was selected, the doors stimulus would disappear, a fixation cross (+) would appear for 1,000 ms, and then feedback was presented on the screen. Feedback was presented either as a green arrow pointing up (↑) signaling a gain of $0.50 or a red arrow pointing down (↓) signaling a loss of $0.25 and was presented for 2,000 ms. A second fixation cross (+) followed the gain or loss feedback offset and was presented for 1,500 ms. Immediately after the second fixation cross, “Click for the next round” was displayed on the screen until the participant began the next trial with a button press.

EEG Recording and Processing

Continuous EEG data were collected using an elastic cap using the ActiveTwo BioSemi system and digitized using ActiView software (BioSemi, Amsterdam, Netherlands). The 34-elctrode sites were placed using the 10/20 system and two external electrodes were placed on the right and left mastoids. To collect electrooculogram (EOG) produced from eye blinks and eye movements, four additional external electrodes were placed on the face: vertical eye movements and blinks were recorded via two electrodes approximately 1 cm above and below the left eye and horizontal movement and eye blinks were recorded via two electrodes placed approximately 1 cm left of the left eye and right of the right eye. EEG data were digitized with a sampling rate of 1024 Hz using a low-pass fifth order sine filter with a half-power cut off of 204.8 Hz. The online reference was created by a common mode sense (CMS) active electrode producing a monopolar (nondifferential) channel, as designed by BioSemi. All bioelectric signals were analyzed offline using Brain Vision Analyzer (Brain Products, Gliching, Germany), re-referenced to the average of the right and left mastoids, and band-passed filtered with low and high cut-offs of 0.01 and 30 Hz with a 24 dB/oct roll-off.

EEG data were segmented at 200 ms pre-feedback and 1,000 ms post feedback (i.e. 1,200 ms epochs). A regression-based approach (Gratton, Coles, & Donchin, 1983) was used on the segments to correct for eye movement artifacts. An automatic artifact rejection procedure was used to detect and reject artifacts for all segmented data. Criteria for rejection is as follows: epochs containing a voltage greater than 50 μV between sample points, a voltage difference of 175 μV within a segment, or a maximum voltage different of less than 0.50 μV within 100 ms intervals. Visual inspection was performed on the segments to identify and remove any remaining artifacts. Finally, the 200 ms pre-stimulus interval was used as the baseline for baseline correction.

Feedback-locked ERPs were averaged separately for gains and losses. The number of trials per condition that remained after artifact rejection at the FCz electrode site were as follows: Gain (M = 19.90, SD = .43), Loss (M = 19.86, SD = .78). The average ERP response to gains and losses between 250 and 350 ms were exported. Finally, the RewP was computed by subtracting the averaged ERP response to monetary losses from the averaged ERP response to monetary gains, thus creating a subtraction-based difference score.

Internal consistency reliability was computed for the neural responses to monetary gains and losses using the Spearman-Brown-corrected correlations, and for the RewP difference score using an adjusted α formula (Furr & Bacharach, 2013). This adjusted α formula is more appropriate for calculating the reliability of difference scores as it accounts for the reliabilities, variances, and intercorrelations of the two contributing measures (i.e., the individual gain and loss responses; Levinson et al., 2017). Spearman-Brown-corrected reliability coefficients for the neural responses to monetary gains and losses were .84 and .77, respectively. Adjusted α for the RewP was .42. This reliability is considered reasonable considering that difference scores inherently have lower reliability (Peter, Churchill, & Brown, 1993; Rogosa, 1995).

Data Analysis

Descriptive statistics were computed for demographics and MPAS scale scores. Next, correlations were utilized to examine associations between MPAS scales and neural responses (i.e., ERP responses to monetary wins and losses, and the RewP). Additionally, a linear regression was conducted to examine associations between MPAS scales and the RewP while controlling for participant age. Finally, a linear regression was conducted to examine associations between the Pleasure in Proximity scale and the RewP while controlling for symptoms of depression, anxiety and stress as measured by the DASS-21 at the baseline assessment.

Results

Descriptives

Descriptive statistics summarizing demographics and MPAS scale scores are presented in Table 1.

Associations between MPAS scales at follow-up

At follow-up, women who reported increased scores on the Pleasure in Proximity subscale also reported increased scores on the Tolerance and Acceptance subscale (r(63) = .42, p = .001).

Associations between reward-related neural activity at baseline and MPAS scales at follow-up

Women with increased (i.e., more positive) neural responses to monetary gains at baseline reported increased Pleasure in Proximity subscale scores at follow-up (r(63) = .25, p = .046). Additionally, women with an increased RewP (i.e., the subtraction-based difference between the neural gain and loss responses) reported increased Pleasure in Proximity subscale scores at follow-up (r(63) = .35, p = .005). The neural response to monetary losses at baseline was unrelated to Pleasure in Proximity scores at follow-up (r(63) = −.02, p = .91). Moreover, baseline neural responses to monetary gains (r(63) = .24, p = .054) and losses (r(63) = .07, p = .58), as well as the baseline RewP (r(63) = .24, p = .059), were all unrelated to Tolerance and Acceptance subscale scores at follow-up. The bivariate relationship between the RewP and Pleasure in Proximity subscale scores is depicted in Figure 1.

Figure 1.

Figure 1.

Open in a new tab

Scatter plot depicting the association between RewP amplitudes and MPAS Pleasure in Proximity subscale scores. The line represents the line of best fit. The present findings were stable when univariate outliers were recoded to be no greater or less than two interquartile ranges from the median.

We also noted that older women showed a reduced neural response to monetary gains at the baseline assessment (r(63) = −.30, p = .019). Thus, we next conducted a linear regression to examine whether the RewP at baseline was associated with Pleasure in Proximity subscale scores at follow-up, independent of age. A linear regression was conducted with the RewP at baseline and age at baseline entered as predictors, and Pleasure in Proximity subscale scores at follow-up entered as the dependent variable. The regression model was significant (F(2, 63) = 5.11, p = .009) with an R2 of .15. Results suggested that the RewP at baseline (B = .19, t = 2.75, p = .008) was a significant predictor of Pleasure in Proximity subscale scores when controlling for age at baseline. However, baseline age was not a significant predictor of Pleasure in Proximity subscale scores (B = −.10, t = −1.28, p = .21). Thus, increased pleasure in proximity with the infant at follow-up was associated with increased RewP amplitude at baseline even when controlling for age at baseline.

To illustrate the association between the RewP measured at the baseline assessment and pleasure in proximity with the infant at follow-up, scores on the Pleasure in Proximity scale were dichotomized using a median split (median = 36.20) and the RewP at baseline was plotted for those with high versus low Pleasure in Proximity scores separately. As evident from Figure 2, women who reported lower pleasure when in proximity to the infant in the postnatal period demonstrated a reduced RewP in pregnancy.

Figure 2.

Figure 2.

Open in a new tab

Antenatal feedback-locked ERPs (left) for gains and losses, and topographic maps for the gain-loss difference (right) in individuals high (top) and low (bottom) in pleasure in proximity in the postnatal period. Individuals with low pleasure in proximity with their infant showed a reduced RewP (i.e., the difference in amplitude between gain and loss conditions) in pregnancy as compared to individuals with high pleasure in proximity with their infant.

Associations between reward-related neural activity, MPAS scales, and depressive and anxious symptoms

Finally, we conducted a linear regression to examine whether the RewP at baseline was associated with Pleasure in Proximity subscale scores at follow-up, independent of the DASS-21 depression, anxiety, and stress scale scores at baseline. A linear regression was conducted with the RewP at baseline and DASS-21 depression, anxiety, and stress scale scores at baseline entered as predictors, and Pleasure in Proximity subscale scores at follow-up entered as the dependent variable. The regression model was significant (F(4, 58) = 3.26, p = .02) with an R2 of .18. Results suggested that the RewP at baseline (B = .23, t = 3.20, p = .002) was a significant predictor of Pleasure in Proximity subscale scores when controlling for DASS-21 scale scores at baseline. However, DASS-21 depression (B = −.26, t = −1.76, p = .083), anxiety (B = .04, t = .32, p = .75), and stress (B = .05, t = .29, p = .77) scale scores at baseline were not significant predictors of Pleasure in Proximity subscale scores. Therefore, increased pleasure in proximity with the infant at follow-up was associated with increased RewP amplitude at baseline even when controlling for depression, anxiety, and stress at the baseline assessment during the antenatal period.

Discussion

The present study examined whether reward-related neural activity in the antenatal period prospectively predicts self-reported mother-to-infant bonding in the postnatal period. Women with an increased neural response to monetary gains and an increased RewP in the antenatal period reported increased pleasure in proximity with their infant at the postnatal follow-up assessment. This association was significant even when controlling for symptoms of depression, anxiety and stress, as well as maternal age, a variable that was negatively associated with the RewP at the baseline assessment. These data suggest that the antenatal RewP was a better predictor of mother-to-infant bonding in the postnatal period than antenatal maternal depression or age.

These findings align with previous research studies that have proposed links between neural reward responsivity and maternal bonding and sensitivity. The RewP is associated with fMRI measures of activity in the ventral striatum in response to monetary reward (Carlson et al., 2011; Becker et al., 2014) and source localization studies have identified the striatum as a likely neural generator for the RewP (Foti et al., 2011). Activation of the ventral striatum has consistently been linked to maternal behavior in rodent models. In rodent studies, researchers have lesioned neural structures that comprise the mesolimbic DA pathway, including the ventral tegmental area and nucleus accumbens, and have shown that these lesions impair rodent maternal behavior, such as nursing (Smith & Holland, 1975), nest building (Gaffori & Le Moal, 1979), and pup retrieval (Gaffori & Le Moal, 1979; Numan & Smith, 1984; Hansen et al., 1991). Furthermore, striatal activation has also been associated with infant cues and attachment quality in human studies. In multiple studies, reward processing regions, such as the ventral striatum, the VTA, and the substantia nigra, have shown greater reactivity to one’s own infant cues in comparison to those of an unknown infant (Strathearn et al., 2008; Strathearn et al., 2009). Furthermore, reduced activation in the ventral striatum in response to images of one’s own baby has been associated with an insecure/dismissing attachment style (Strathearn et al., 2009). Thus, the present findings align with this previous work and provide further evidence for the importance of reward-related neural activity for maternal bonding and mother-to-infant attachment.

The present findings also extend on prior research in a number of ways. First, the present study employed monetary rewards, in contrast to many studies of maternal sensitivity and attachment, which have commonly utilized infant-related stimuli. The presence of associations between maternal bonding and the neural response to monetary rewards suggests that broader reward processing deficits may impact a mother’s bond with her infant, and not just deficits to infant-related cues. This aligns with work on addiction and depression—these conditions are associated with overall dampening of activity in reward circuits to a wide range of stimuli, and indeed, maternal addiction and depression are associated with dampened striatal responses to infant cues (Kim et al., 2018; Laurent & Ablow, 2012) and with reduced mother-to-infant attachment (Moses-Kolko et al., 2011). As such, it can be suggested that neural responding to monetary rewards, as an indicator of general reward responsivity, may predict the degree of pleasure experienced in bonding with, and caring for, the infant. This also raises the possibility that increased pleasure in proximity with the infant may very well be reflective of one’s general tendency to experience pleasure. Second, the present findings are novel in that they suggest the presence of a prospective association between neural reward sensitivity and self-reported attachment quality, which has yet to have been investigated. Thus, our findings raise the intriguing possibility that reduced reward responsivity in pregnancy could function to identify individuals that may benefit from early intervention, and that interventions aiming to increase neural reward responsiveness may result in improved mother-to-infant bonding and attachment outcomes.

The present study has limitations that warrant consideration. Specifically, generalizability of these findings is limited as the majority of our participants were Caucasian women. Additionally, as this is the first study to examine whether the neural response to monetary reward in pregnancy predicts mother-infant bonding in the postnatal period, this study provides preliminary evidence of potential link between general reward sensitivity in pregnancy and maternal bonding in the postnatal period that should be further investigated and replicated.

In conclusion, the current study examined whether EEG measures of reward responsiveness in pregnancy may predict postnatal self-report measures of mother-to-infant attachment. Results revealed that an increased RewP measured in pregnancy was associated with increased pleasure in proximity with the infant measured at approximately one year postpartum. Furthermore, regression analyses revealed that this association was independent of symptoms of depression, anxiety, and stress. This is the first study to examine reward function as a prospective predictor of maternal bonding and attachment to the infant. As such, our findings set the stage for future studies examining whether reduced reward responsiveness in pregnancy may aid in the early detection of risk for reduced mother-infant attachment, and whether interventions aiming to increase neural reward responsivity may result in improved attachment outcomes.

Acknowledgments:

We would like to thank Bill Hambsh, David O’Bryan, Erin Ryals, and the staff at North Florida Women’s Care for their support with subject recruitment.

Funding:

This work was supported by the following grants: NIMH R21MH116481, NIMH T32MH093311, NIMH F31MH125624

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Declarations of interest: None.

1

We compared age, gestational weeks at baseline, and DASS-21 scores at baseline between participants who did and did not provide follow-up data. While trending, age did not vary between participants who did and did not provide follow-up data, t(235) = −1.94, p = .053. Furthermore, gestational weeks at baseline (t(234) = −1.74, p = .08) and DASS-21 depression (t(235) = 1.62, p = .11), anxiety (t(235) = 1.50, p = .14), and stress (t(235) = 1.70, p = .09) scores also did not vary between groups.

References

  1. Ainsworth MS (1979). Infant-mother attachment. American Psychologist, 34(10), 932. [DOI] [PubMed] [Google Scholar]
  2. Alhusen JL, Hayat MJ, & Gross D (2013). A longitudinal study of maternal attachment and infant developmental outcomes. Archives of women’s mental health, 16(6), 521–529. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Antony MM, Bieling PJ, Cox BJ, Enns MW, & Swinson RP (1998). Psychometric properties of the 42-item and 21-item versions of the Depression Anxiety Stress Scales in clinical groups and a community sample. Psychological assessment, 10(2), 176. [Google Scholar]
  4. Atkinson L, Paglia A, Coolbear J, Niccols A, Parker KC, & Guger S (2000). Attachment security: A meta-analysis of maternal mental health correlates. Clinical psychology review, 20(8), 1019–1040. [DOI] [PubMed] [Google Scholar]
  5. Baker TE, & Holroyd CB (2011). Dissociated roles of the anterior cingulate cortex in reward and conflict processing as revealed by the feedback error-related negativity and N200. Biological psychology, 57(1), 25–34. [DOI] [PubMed] [Google Scholar]
  6. Becker MP, Nitsch AM, Miltner WH, & Straube T (2014). A single-trial estimation of the feedback-related negativity and its relation to BOLD responses in a time-estimation task. Journal of Neuroscience, 34(8), 3005–3012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bress JN, & Hajcak G (2013). Self-report and behavioral measures of reward sensitivity predict the feedback negativity. Psychophysiology, 50(7), 610–616. [DOI] [PubMed] [Google Scholar]
  8. Broad KD, Curley JP, & Keverne EB (2006). Mother-infant bonding and the evolution of mammalian social relationships. Philosophical Transactions of the Royal Society B: Biological Sciences, 361(1476), 2199–2214. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Carlson JM, Foti D, Mujica-Parodi LR, Harmon-Jones E, & Hajcak G (2011). Ventral striatal and medial prefrontal BOLD activation is correlated with reward-related electrocortical activity: a combined ERP and fMRI study. Neuroimage, 57(4), 1608–1616. [DOI] [PubMed] [Google Scholar]
  10. Champagne FA, Chretien P, Stevenson CW, Zhang TY, Gratton A, & Meaney MJ (2004). Variations in nucleus accumbens dopamine associated with individual differences in maternal behavior in the rat. Journal of Neuroscience, 24(17), 4113–4123. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Condon JT, & Corkindale CJ (1998). The assessment of parent-to-infant attachment: Development of a self-report questionnaire instrument. Journal of Reproductive and Infant Psychology, 16(1), 57–76. [Google Scholar]
  12. De Carli P, Costantini F, Sessa P, Visentin S, Pearson RM, & Simonelli A (2019). The expectant social mind: A systematic review of face processing during pregnancy and the effect of depression and anxiety. Neuroscience & Biobehavioral Reviews, 102, 153–171. [DOI] [PubMed] [Google Scholar]
  13. DeKlyen M, & Greenberg MT (2008). Attachment and psychopathology in childhood. In Cassidy J & Shaver PR (Eds.), Handbook of attachment: Theory, research, and clinical applications (2nd ed., pp. 637–665). New York, NY: Guilford. [Google Scholar]
  14. Dudek J, Colasante T, Zuffianò A, & Haley DW (2020). Changes in cortical sensitivity to infant facial cues from pregnancy to motherhood predict mother-infant bonding. Child development, 91(1), el98–e217. [DOI] [PubMed] [Google Scholar]
  15. Foti D, Weinberg A, Dien J, & Hajcak G (2011). Event-related potential activity in the basal ganglia differentiates rewards from nonrewards: Temporospatial principal components analysis and source localization of the feedback negativity. Human brain mapping, 32(12), 2207–2216. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Furr RM, & Bacharach VR (2013). Psychometrics: An introduction (2nd ed.). Thousand Oaks, CA: Sage Publications [Google Scholar]
  17. Gaffori O, & Le Moal M (1979). Disruption of maternal behavior and appearance of cannibalism after ventral mesencephalic tegmentum lesions. Physiology & Behavior, 23(2), 317–323. [DOI] [PubMed] [Google Scholar]
  18. Gholampour F, Riem MM, & van den Heuvel MI (2020). Maternal brain in the process of maternal-infant bonding: Review of the literature. Social Neuroscience, 15(4), 380–384. [DOI] [PubMed] [Google Scholar]
  19. Gratton G, Coles MG, & Donchin E (1983). A new method for off-line removal of ocular artifact. Electroencephalography and clinical neurophysiology, 55(4), 468–484. [DOI] [PubMed] [Google Scholar]
  20. Groh AM, Fearon RP, Bakermans-Kranenburg MJ, Van IJzendoorn MH, Steele RD, & Roisman GI (2014). The significance of attachment security for children’s social competence with peers: A meta-analytic study. Attachment & human development, 16(2), 103–136. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Groh AM, Fearon RP, van IJzendoorn MH, Bakermans-Kranenburg MJ, & Roisman GI (2017). Attachment in the early life course: Meta-analytic evidence for its role in socioemotional development. Child Development Perspectives, 11(1), 70–76. [Google Scholar]
  22. Groh AM, Roisman GI, van IJzendoorn MH, Bakermans-Kranenburg MJ, & Fearon RP (2012). The significance of insecure and disorganized attachment for children’s internalizing symptoms: A meta-analytic study. Child development, 83(2), 591–610. [DOI] [PubMed] [Google Scholar]
  23. Hajcak G, Moser JS, Holroyd CB, & Simons RF (2007). It’s worse than you thought: The feedback negativity and violations of reward prediction in gambling tasks. Psychophysiology, 44(6), 905–912. [DOI] [PubMed] [Google Scholar]
  24. Hansen S, Harthon C, Wallin E, Löfberg L, & Svensson K (1991). Mesotelencephalic dopamine system and reproductive behavior in the female rat: effects of ventral tegmental 6-hydroxydopamine lesions on maternal and sexual responsiveness. Behavioral neuroscience, 105(4), 588. [DOI] [PubMed] [Google Scholar]
  25. Hoekzema E, Barba-Müller E, Pozzobon C, Picado M, Lucco F, García-García D, … & Vilarroya O (2017). Pregnancy leads to long-lasting changes in human brain structure. Nature neuroscience, 20(2), 287–296. [DOI] [PubMed] [Google Scholar]
  26. Hoekzema E, Tamnes CK, Berns P, Barba-Müller E, Pozzobon C, Picado M, … & Carmona S (2020). Becoming a mother entails anatomical changes in the ventral striatum of the human brain that facilitate its responsiveness to offspring cues. Psychoneuroendocrinology, 112, 104507. [DOI] [PubMed] [Google Scholar]
  27. Izuma K, Saito DN, & Sadato N (2008). Processing of social and monetary rewards in the human striatum. Neuron, 58(2), 284–294. [DOI] [PubMed] [Google Scholar]
  28. Jacobvitz D, & Hazen N (1999). Developmental pathways from infant disorganization to childhood peer relationships. In: Solomon J, George C, eds. Attachment Disorganization. New York: Guilford Press, 1999. [Google Scholar]
  29. Kim S, Iyengar U, Mayes LC, Potenza MN, Rutherford HJ, & Strathearn L (2017). Mothers with substance addictions show reduced reward responses when viewing their own infant’s face. Human Brain Mapping, 38(11), 5421–5439. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Laurent HK, & Ablow JC (2012). A cry in the dark: depressed mothers show reduced neural activation to their own infant’s cry. Social cognitive and affective neuroscience, 7(2), 125–134. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Levinson AR, Speed BC, Infantolino ZP, & Hajcak G (2017). Reliability of the electrocortical response to gains and losses in the doors task. Psychophysiology, 54(4), 601–607. [DOI] [PubMed] [Google Scholar]
  32. Lewis M, & Feiring C (1989). Infant, mother, and mother-infant interaction behavior and subsequent attachment. Child development, 831–837. [Google Scholar]
  33. Lin A, Adolphs R, & Rangel A (2012). Social and monetary reward learning engage overlapping neural substrates. Social cognitive and affective neuroscience, 7(3), 274–281. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Lindgren K (2001). Relationships among maternal–fetal attachment, prenatal depression, and health practices in pregnancy. Research in nursing & health, 24(3), 203–217. [DOI] [PubMed] [Google Scholar]
  35. Lovibond PF, & Lovibond SH (1995). The structure of negative emotional states: Comparison of the Depression Anxiety Stress Scales (DASS) with the Beck Depression and Anxiety Inventories. Behaviour research and therapy, 33(3), 335–343. [DOI] [PubMed] [Google Scholar]
  36. Moses-Kolko EL, Fraser D, Wisner KL, James JA, Saul AT, Fiez JA, & Phillips ML (2011). Rapid habituation of ventral striatal response to reward receipt in postpartum depression. Biological psychiatry, 70(4), 395–399. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Numan M (2007). Motivational systems and the neural circuitry of maternal behavior in the rat. Developmental Psychobiology: The Journal of the International Society for Developmental Psychobiology, 49(1), 12–21. [DOI] [PubMed] [Google Scholar]
  38. Numan M, & Smith HG (1984). Maternal behavior in rats: evidence for the involvement of preoptic projections to the ventral tegmental area. Behavioral neuroscience, 98(4), 712. [DOI] [PubMed] [Google Scholar]
  39. Olazabal DE, & Young LJ (2006). Oxytocin receptors in the nucleus accumbens facilitate “spontaneous” maternal behavior in adult female prairie voles. Neuroscience, 141(2), 559–568. [DOI] [PubMed] [Google Scholar]
  40. Pearson RM, Lightman SL, & Evans J (2011). Attentional processing of infant emotion during late pregnancy and mother–infant relations after birth. Archives of women’s mental health, 14(1), 23–31. [DOI] [PubMed] [Google Scholar]
  41. Peter JP, Churchill GA Jr, & Brown TJ (1993). Caution in the use of difference scores in consumer research. Journal of consumer research, 19(4), 655–662. [Google Scholar]
  42. Proudfit GH (2015). The reward positivity: From basic research on reward to a biomarker for depression. Psychophysiology, 52(4), 449–459. [DOI] [PubMed] [Google Scholar]
  43. Rincón-Cortés M, & Grace AA (2020). Adaptations in reward-related behaviors and mesolimbic dopamine function during motherhood and the postpartum period. Frontiers in Neuroendocrinology, 100839. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Rogosa D (1995). Myths and methods:“Myths about longitudinal research” plus supplemental questions. The Analysis of Change, 3, 66. [Google Scholar]
  45. Smith MO, & Holland RC (1975). Effects of lesions of the nucleus accumbens on lactation and postpartum behavior. Physiological psychology, 3(4), 331–336. [Google Scholar]
  46. Sroufe LA (2005). Attachment and development: A prospective, longitudinal study from birth to adulthood. Attachment & human development, 7(4), 349–367. [DOI] [PubMed] [Google Scholar]
  47. Sroufe LA, Egeland B, Carlson E, & Collins WA (2005). Placing Early Attachment Experiences in Developmental Context: The Minnesota Longitudinal Study. In Grossmann KE, Grossmann K, & Waters E (Eds.), Attachment from infancy to adulthood: The major longitudinal studies (p. 48–70). Guilford Publications. [Google Scholar]
  48. Strathearn L, Fonagy P, Amico J, & Montague PR (2009). Adult attachment predicts maternal brain and oxytocin response to infant cues. Neuropsychopharmacology, 34(13), 2655–2666. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Strathearn L, Li J, Fonagy P, & Montague PR (2008). What’s in a smile? Maternal brain responses to infant facial cues. Pediatrics, 122(1), 40–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Swain JE, Kim P, Spicer J, Ho SS, Dayton CJ, Elmadih A, & Abel KM (2014). Approaching the biology of human parental attachment: Brain imaging, oxytocin and coordinated assessments of mothers and fathers. Brain research, 1580, 78–101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Tversky A, & Kahneman D (1981). The framing of decisions and the psychology of choice. Science, 211(4481), 453–458. [DOI] [PubMed] [Google Scholar]
  52. Tversky A, & Kahneman D (1992). Advances in prospect theory: Cumulative representation of uncertainty. Journal of Risk and uncertainty, 5(4), 297–323. [Google Scholar]

RESOURCES