Mother–infant bonding in the first nine months postpartum: the role of mother’s attachment style and psychological flexibility

ABSTRACT

Introduction

Mother’s bond to the infant in the postpartum period plays an important role in the subsequent mother–infant relationship and the infant’s socio-emotional functioning. Several maternal characteristics, such as attachment style and psychological flexibility, may contribute to the quality of mother–infant bonding, though literature examining these variables is still scarce. The present study aimed to examine the impact of mother’s attachment on mother–infant bonding in the first month postpartum and the mediating role of psychological flexibility on that association.

Methods

Participants were 226 mothers of an infant up to 9 months old, who reported on their own attachment style (in terms of anxiety, comfort with proximity, trust in others), psychological flexibility (in terms of openness to experience, behavioural awareness, valued action) and mother–infant bonding.

Results

Results showed that mother’s attachment anxiety predicted a bond with the infant directly and indirectly via mother’s psychological flexibility, specifically through behavioural awareness and valued action. Trust in others had an impact on mother–infant bonding through behavioural awareness, whereas comfort with proximity influenced mother–infant bond indirectly, via valued action. Finally, mothers’ civil status, schooling and number of children were relevant to better understand the variance of our mediating and dependent variables.

Discussion

Our findings highlight the importance of mother’s attachment and psychological flexibility in promoting the quality of mother–infant bonding, which can inform future intervention programmes targeting modifiable factors, such as psychological flexibility, to promote early positive parent–infant relationships, particularly for single, first-time mothers, with higher levels of education.

KEYWORDS:

• Mother–infant bonding
• mother’s attachment
• psychological flexibility
• postpartum period
• mediation analysis

Introduction

Mother–infant bonding, typically characterised by mother’s feelings and emotions towards her child, has its onset still during pregnancy, develops primarily during the first postpartum year and may continue throughout the child’s life (de Cock et al., 2016; Kinsey & Hupcey, 2013). Specific parental behaviours are often used to bond to the infant, such as seeking proximity, touch, eye contact, baby talk, smiling and adaptation to the infant’s cues (Reck et al., 2006). This emotional tie of a mother to her baby will ensure that the infant receives the protection and care needed, promotes the quality of mother–infant relationship, and contributes to positive infant outcomes. Specifically, high quality of bonding has been related to more maternal sensitivity and involvement during the interaction with their infant (e.g. Maas et al., 2016;) and to more positive child socio-emotional functioning (de Cock et al., 2016; Meins et al., 2018; Winston & Chicot, 2016). However, not all women develop an optimal mother–infant bonding. Among other factors, this process may be influenced by several parental characteristics, such as mental health (e.g. depression, anxiety) (de Cock et al., 2016; Nonnenmacher et al., 2016; Tichelman et al., 2019), personality (de Cock et al., 2016), emotional regulation abilities (Sahin, 2022) and attachment representations (Nonnenmacher et al., 2016).

One aspect that has been associated with bonding is the mother’s own attachment. Adult’s representation of early relational experiences with their caregivers shapes the way they will establish romantic and other significant interpersonal relationships (Antonucci et al., 2004; Waters et al., 2000), as well as the quality of mother–infant bonding and parenting behaviours towards the child (e.g. Alhusen et al., 2013; Jones et al., 2015; Tichelman et al., 2019). Specifically, avoidant or anxious attachment styles have been related to less sensitive, responsive and supportive behaviour when interacting with the child (Berlin et al., 2011; Jones et al., 2015; Selcuk et al., 2010). Those attachment styles were also found to be associated with parenting stress and poorer mother–infant bonding at 8 weeks postpartum (Nordahl et al., 2020), with that relationship being further mediated by parenting stress. Similarly, further studies observed that mothers with insecure (anxious and avoidant) attachment styles reported bonding difficulties to their child and this association was mediated by maternal psychopathology (Hairston et al., 2018; Handelzalts et al., 2021). Research shows that individuals with insecure attachment styles tend to present maladaptive emotion regulation skills – such as self-blame and rumination – that increase the risk for postpartum depression, which, in turn, has been consistently found to predict difficulties in mother–infant outcomes (Brake et al., 2020; Haga et al., 2012). The mother’s responsiveness towards the child’s needs relies on mothers’ emotion regulation abilities (Morris et al., 2017), as they need to be able to effectively regulate both their own and their child’s emotional state (Morelen et al., 2016; Rutherford et al., 2015).

Considering that poorer bonding can lead to adverse outcomes for the mother (Maas et al., 2016) and the child (e.g. Winston & Chicot, 2016), it is important to investigate which factors relate to maternal-infant bonding, especially those that can be modified through targeted interventions. One relevant potential mediator to be explored is psychological flexibility (PF), a core component of the Acceptance and Commitment Therapy (ACT; Cherry et al., 2021; Hayes et al., 2012). PF refers to how the individual relates more fully and consciously to all their present experiences and adopts behaviours that serve their personal values. In contrast, psychological inflexibility reflects rigid patterns of responding to the present moment, by attempting to control or avoid unpleasant internal experiences and adopting behaviours that do not align with one’s values (Cherry et al., 2021; Hayes et al., 2006). Psychological (in)flexibility has been related to psychopathology, emotion regulation abilities and functioning across many life domains (Cherry et al., 2021; Daks & Rogge, 2020; Doorley et al., 2020). Recently, a meta-analysis (Daks & Rogge, 2020) concluded that psychological (in)flexibility plays a key role in how individuals interact within one’s family and related to more adaptive parenting strategies, lower levels of parental stress, increased family cohesion and child well-being.

To our knowledge, studies investigating the mediating role of PF in the association between adult’s attachment style and mother–infant bonding are still lacking, although recent research has examined the associations between those variables separately. For example, in a study with mothers of children with autism spectrum disorder, maternal avoidant and anxious attachment were correlated with greater psychological inflexibility (Kulasinghe et al., 2022). Also, psychological flexibility has been shown to mediate the association between adult attachment style and individuals’ psychological well-being, such that anxious and avoidant attachment styles negatively impacted psychological well-being through increased levels of psychological inflexibility (Calvo et al., 2022; Li et al., 2022). In what concerns mother–infant bonding, Evans et al. (2012) also showed that higher levels of mother’s psychological flexibility were associated with greater maternal responsiveness to the infant’s signs and a stronger bond with the infant. Furthermore, Whittingham and Mitchell (2021) observed that maternal psychological flexibility significantly predicted mother’s perception of emotional availability in the relationship with their child, assessed in terms of mutual attunement and affect quality.

Considering this (albeit partialised) evidence, the present study aimed to investigate the contribution of mothers’ attachment and psychological flexibility to mother–infant bonding and whether the mother’s psychological flexibility can mediate the relationship between the mother’s attachment style and her bond to the infant. We hypothesise that mother’s attachment characterised by less anxiety, more trust in others and more comfort with proximity to others will impact on stronger mother–infant bonding, as will greater psychological flexibility. Moreover, we expect psychological flexibility to mediate the impact of mothers’ attachment on mother–infant bonding.

Methods

Participants

The sample consisted of 226 women, aged between 19 and 42 years (M = 32.05; SD = 4.35), who had an infant up to 9 months of age. For most of these women, this infant was their first child (74.8%, n = 169). Most participants were married or living in a de facto partnership (n = 204; 90.3%) and had college degree (n = 117, 78.3%). Most participants had only one child (n = 169, 74.8%), 50 participants had 2 children (2,1%) and 7 participants (3,1%) had between 3 and 5 children. Infants were on average 3.69 months old (SD = 2.23), ranging from 0 to 9 months.

Inclusion criteria to participate in the study were a) being 18-year-old or older; b) having an infant aged 9 months or younger and c) being fluent in Portuguese language. In turn, being a mother of twins was an exclusion criterion.

Measures

Socio-demographic information. Participants completed socio-demographic questions to collect information about their age, educational level, marital status, number of children, previous pregnancies and infant age.

Mother Attachment

Mothers’ attachment was assessed through the Revised Adult Attachment Scale (AAS-R; Collins, 1996, Portuguese version by Canavarro et al., 2006), which is composed of 18 items answered on a 5-point scale ranging from 1 = ‘Not at all characteristic of me’ to 5 = ‘Extremely characteristic of me’, based on how they feel about their (past and present) romantic relationships. The items are organised in three subscales, described in the Portuguese version of the AAS-R as follows: anxiety (level of anxiety that the individual feels about interpersonal issues such as fear of abandonment or not being well-liked); comfort with proximity (degree to which the individual feels comfortable with proximity and intimacy) and trust in others (degree of trust in others and in their availability, when necessary) (Canavarro et al., 2006). Some of the items are reverse scored so that higher scores will reflect greater anxiety, more comfort with proximity and more trust in others. The Portuguese version of the AAS-R presented a Cronbach's alpha of .84 for anxiety, .67 for contact with proximity and .54 for trust in other subscales (Canavarro et al., 2006). In our sample, the Cronbach’s alphas were .75, .63 and .86, respectively.

Psychological flexibility

The Comprehensive Assessment of Acceptance and Commitment Therapy Processes (CompACT; Francis et al., 2016; Portuguese version by Trindade et al., 2021) was used to measure the mother’s psychological flexibility. This self-report questionnaire consists of 18 items scored on a 7-point Likert scale, which ranges from 0 = ‘Strongly disagree’ to 6 = ‘Strongly agree’. Three subscales emerge from the items: openness to experience (willingness to be in contact with one’s inner experiences, including unpleasant ones, without trying to control or avoid them); behavioural awareness (one’s ability to be fully aware of what happens in the present moment instead of engaging in automatic responses) and valued action (one’s willingness to act and live according to ones’ own personal values) (Trindade et al., 2022). Higher scores on openness to experience and behavioural awareness subscales indicate greater psychological inflexibility, whereas higher scores on the valued action subscale reflect higher levels of psychological flexibility. The Portuguese version of CompACT presented acceptable to good internal consistencies (openness to experience: α = .77; behavioural awareness: α = .87; valued action: α = .86) (Trindade et al., 2021). In our sample, the Cronbach’s alphas were .71, .90 and .86, respectively.

Mother–infant bonding

The Parent Bonding Questionnaire (PBQ; Brockington et al., 2006, Portuguese version by Nazaré et al., 2012) is a self-report instrument used to measure mother–infant bonding in the postpartum period. The instrument consists of 12 items responded in a 6-point Likert scale, ranging from 0 = ‘Never’ to 5 = ‘Always’. The PBQ identifies one single dimension labelled as impaired bonding (e.g. Nazaré et al., 2012). Items with negative valence were reverse coded, such that higher scores would reflect better quality of mother–infant bonding. The internal consistency revealed to be adequate either in the study by Nazaré et al. (2012; Cronbach’s α = 0.71) and in our sample (Cronbach’s α = 0.77).

Procedure

Data were collected through an online survey between October 2021 and January 2022. Participants’ recruitment took place through social media advertising and through acquaintances of the research team. The questionnaires were distributed via a Google Forms link, which provided a description of the main objectives and informed participants about the voluntary character of their participation and the confidential and anonymous nature of the collected data. Only after giving their informed consent, by clicking the accept to participate button, the participants had full access to the study questionnaires.

Data analysis

Preliminary data analysis was performed using IBM SPSS Statistics, version 24.0 (IBM Corporation, 2016). Descriptive statistics of sample characteristics and study target variables were presented in terms of frequencies and percentages for categorical variables, whereas the mean and standard deviation were reported for continuous variables. The correlation between the mother’s attachment and psychological flexibility subscales and mother–infant bonding was also examined. Finally, mediation analysis was run through Mplus version 7.1 (Muthén & Muthén, 1998-2017). The model examined the direct effect of mother’s attachment dimensions and psychological flexibility on mother–infant bonding and explored whether the effect of mother’s attachment dimensions on mother–infant bonding was mediated by measures of psychological flexibility. Model fit was assessed considering a Comparative Fit Index (CFI) higher than .97 and Root Mean Square Error of Approximation (RMSEA) lower than .08 (Hair et al., 2014). The explained variance of the dependent and mediator variables was considered. To ascertain for the association between sociodemographic variables and the variance of the dependent variable explained by the model, those sociodemographic variables (i.e. mothers’ age, civil status, and schooling, number of children and age of the youngest child) were then entered into the model as predictors of both the mediating and the dependent variables.

Results

Preliminary descriptive and correlation analyses

The descriptive statistics of study variables are presented in Table 1. Pearson correlation analyses further showed that mother’s attachment was significantly related to mother–infant bonding (see Table 1). More specifically, greater anxiety was negatively associated with quality of mother–infant bonding, whereas higher levels of trust in others and comfort with proximity were associated with better quality of mother–infant bonding. Similarly, all psychological flexibility subscales yielded significant associations with mother–infant bonding (see Table 1). Thus, a positive mother–infant bonding was related to higher levels of valued action (reflecting greater psychological flexibility), and to lower levels of openness to experience and behavioural awareness, both indicating more psychological inflexibility through higher scores on the subscales.

Table 1. Descriptive statistics of study variables and Pearson correlation between mother’s attachment and psychological flexibility and mother–infant bonding.

	Min–Max	M (SD)	Mother–Infant Bonding
Mother’s Attachment
Anxiety	6–29	14.62 (5.93)	-.34***
Comfort with proximity	7–30	23.38 (3.98)	.24***
Trust in others	7–26	18.72 (4.14)	.29***
Mother’s Psychological Flexibility
Openness to experience	1–30	17.26 (5.99)	-.24***
Behavioural awareness	0–30	14.11 (7.28)	-.34***
Valued action	6–48	36.54 (6.86)	.31***
Mother–Infant Bonding	36–60	54.13 (5.32)	-

Note: For openness to experience and behavioural awareness, higher scores represent psychological inflexibility, whereas for valued action higher scores represent psychological flexibility; ***p < .001.

Mediation model

A mediation analysis was run for the base model displayed in Figure 1. However, fit indices for this model were not adequate, χ²(3) = 69.75, p = < .001; RMSEA = 0.31; IC = 0.25, 0.38; CFI = 0.74; SRMR = 0.07. To obtain a more parsimonious model, we retained only the statistically significant paths for further analysis. Non-significant paths were removed from the model in the following order: openness to experience predicting mother–infant bonding (p = .974), and, therefore, paths between maternal attachment subscales and openness to experience were removed as they would not contribute to mother–infant bonding; comfort with proximity predicting mother–infant bonding (p = .883); comfort with proximity predicting behavioural awareness (p = .693); trust in others predicting openness to experience (p = .503) and trust in others predicting mother–infant bonding (p = .355).

Figure 1. Base model tested in the mediation analysis to predict mother–infant bonding.

The rectangles with the study variables are linked by straight lines, showing the association between all of them.

The simplified final model (see Figure 2) explained 17% of variation in mother–infant bonding and was a good fit to the data, with the following fit indices: χ²(5) = 8.66, p = .13; RMSEA = 0.06; IC = 0.00, 0.12; CFI = 0.97; SRMR = 0.04. Subscales of maternal attachment significantly explained mother–infant bonding either directly or indirectly through maternal psychological flexibility.

Figure 2. Simplified final pathway model predicting mother–infant bonding. Only significant paths were retained in the model.

Note: The scheme presents only the study variables, represented as rectangles and linked by straight lines, which emerged as statistically related to the mother–infant bonding.

More specifically, mother’s higher anxiety levels in attachment were directly related to lower quality of bond with the infant. Also, two indirect effects were found, with higher levels of valued action (β = −.07, p = .01) and lower levels of behavioural awareness (β = −.18, p = .013), mediating the relationship between mothers’ anxiety in attachment and mother–infant bonding. Another indirect path was observed through the effect of trust in others on mother’s behavioural awareness, such that lower levels of trust in others were associated with less behavioural awareness (greater psychological inflexibility), which had a negative impact on mother–infant bonding (β = .04, p = .05). Additionally, mother’s comfort with proximity explained higher levels of valued action (greater psychological flexibility), which, in turn, revealed an effect on the quality of mother–infant bonding that was almost statistically significant (β = .06, p = .06). No significant paths were found for the effect of openness to experience to mother–infant bonding.

To ascertain for the association between sociodemographic variables and the variance of the dependent variable explained by the model, those variables were considered into this modified model. After deleting non-significant pathways, the adjusted model was an acceptable fit for the data (i.e. χ²(9) = 19.53, p = .06; RMSEA = 0.061; IC = 0.000, 0.106; CFI = 0.97; SRMR = 0.030). In addition to the pathways defined for the modified model that were similar relating to their direction, magnitude and significance level (see Figure 3), there was a significant association between (1) civil status and the (lack of) behavioural awareness, valued action and mother infant bonding, (2) between schooling and the quality of mother–baby bonding and (3) between number of children and (lack of) behavioural awareness. About civil status, single mothers tendentially reported lower quality of bonding to their baby and of both (lack of) behavioural awareness and less valued action, in comparison with mothers who were married or living in a de facto partnership had. Concerning schooling, mothers who had a college degree perceived lower quality of bonding to their baby, in comparison with mothers who had completed the 12^th grade. In relation to number of children, having more children was related to lower levels of psychological inflexibility, as reflected in higher levels of behavioural awareness.

Figure 3. Simplified adjusted pathway model predicting mother–infant bonding. The contribution of sociodemographic variables is presented in dashed lines. Civil status is coded as 1 = single or divorced and 2 = married or in a de facto relationship. Schooling is coded as 1 = finished High School and 2 = has a Higher Education degree. Number of children varied between 1 and 5 children.

Note: The scheme presents the study variables that were statistically associated with mother–infant bonding, represented as rectangles and linked by straight lines, and adds the contribution of sociodemographic variables (civil status, schooling and number of children) presented in dashed lines.

Discussion

The importance of adult’s representation of their attachment style is key to the development of other significant interpersonal relationships, including mother–infant bond (Tichelman et al., 2019), which, in turn, has an impact on both mother and infant outcomes (Maas et al., 2016; Winston & Chicot, 2016). However, research on the role of mothers’ internal psychological processes, namely psychological flexibility, as an outcome of those attachment styles and a precursor to the quality of mother–infant bonding, is still lacking. Thus, this study investigated the predictive value of mother’s attachment and psychological flexibility to mother–infant bond, as well as the role played by psychological flexibility in mediating the effect between adult attachment style and bonding. By examining maternal variables that are likely to be associated with bonding and potentially modified through targeted interventions, such as psychological flexibility, our findings may contribute to the design of early parenting programmes that ain to promote positive parent–infant relationships.

Results revealed that attachment characterised by lower levels of anxiety directly influenced higher quality of mother–infant bonding, which corroborates previous literature (Hairston et al., 2018; Handelzalts et al., 2021; Nordahl et al., 2020; Van Bussel et al., 2010) suggesting that adult representation of one’s attachment in close relationships may also impact their parental role, and especially how they interact and bond to their infant. Individuals with a more anxious and preoccupied attachment orientation tend to report less sensitivity and cooperation and higher levels of overinvolvement and closeness to others, suggesting more inconsistent, controlling and intrusive forms of caregiving (Feeney & Collins, 2001). Selcuk et al. (2010) also found that mother’s attachment-related anxiety was associated with specific parenting behaviours, such as missing the infant’s signs, limiting the infant’s exploration of the environment and more conflict in parent–infant interactions. Therefore, this pattern of attachment is likely to affect the mother’s ability to be sensitive, responsive and engaged with her infant, which is linked to the mother’s emotional regulation skills (Morris et al., 2017). However, these maternal abilities are key parental characteristics to promote early mother–infant bonding (e.g. Brake et al., 2020; Maas et al., 2016).

Our work adds to previous knowledge suggesting that mothers’ abilities to be attuned with their present-moment experiences and being able to behave in accordance with their own life values – as facets of PF – mediate the relationship between mother’s attachment anxiety and the quality of the bond. Previous studies have highlighted the association between psychological functioning, well-being and caregiving (e.g. de Cock et al., 2016; Nonnenmacher et al., 2016; Tichelman et al., 2019). A more anxious orientation in relationships has been related to greater emotional distress and negative affect (Brake et al., 2020; Collins, 1996), which can negatively impact the individual’s ability to change their behaviour and pursue their goals, as maladaptive psychological reactions will prevail and guide their acts, again reducing their psychological well-being (Calvo et al., 2022; Evans et al., 2012). Moreover, PF has been suggested as an important internal psychological competency, key to adaptive emotion regulation, and associated with the way mothers are attuned and responsive towards their babies, as well as with with lower levels of parenting stress (e.g. Fonseca et al., 2020; Whittingham & Mitchell, 2021). Our data are in line with these findings, suggesting that the relation between mothers’ anxiety attachment style and their mother–infant bonding experiences seems to occur partially through mothers’ difficulties in being mindfully aware of their behaviours (in other words, being trapped in the automatic pilot) and through the tendency to engage in behaviours that are not in line with their own values.

We also found that a stronger mother–infant bonding was indirectly influenced by mothers’ higher levels of trust in their attachment relationships with others, through its effect on their PF, reflected by higher levels of behavioural awareness. Furthermore, and even though the indirect effect failed to reach statistical significance, results seem to point that mothers’ higher levels of comfort with proximity in their relationships with others may play a role in the quality of their bonding through mothers’ tendency to maintain behaviours reflecting their values. These findings align with previous works suggesting that greater psychological flexibility mayhap increase the involvement in more effective and positive parenting practices, thus promoting the quality of early bond with the infant (e.g. Brassell et al., 2016; Evans et al., 2012). Thus, mothers who show greater behavioural awareness and valued actions may be more sensitive to the infant’s cues and willing to meet their needs and respond to them in a contingent and appropriate way. These mothers can also be more confident in their role as a parent and have more realistic expectations and beliefs about caregiving practices, thus being more likely to act on purpose towards their life values.

Interestingly, openness to experience did not reveal a direct or indirect effect on mother–infant bonding. In this regard, a recent study using a mixed sample of community adults and postpartum women found different patterns of association between the specific psychological flexibility processes and various domains of the individual’s functioning (e.g. mental health, self-compassion; Trindade et al., 2022). This study observed that postpartum women scored lower than community adult women only in openness to experience. The authors speculated that the postpartum period may be characterised by attempts to (adaptively) avoid and control unwanted thoughts and emotions, as well as unpleasant experiences regarding motherhood, such as extreme tiredness, sleep deprivation and self-criticism about one’s competence as a mother, which could explain the lower scores in the openness to experience scale observed among postpartum women. This may also be why this component was not associated with bonding with the infant: mothers focus on acting in ways that maintain the well-being of their child, which is in accordance with their values, irrespective of being less open to (negative) experiences that may also be taking place.

Findings discussed so far were evident regardless of the specificities of sociodemographic variables. Nevertheless, civil status, schooling and number of children were relevant to better understand the variance of our mediating and dependent variables. Mothers who had a significant other (likely a coparent) reported more psychological flexibility and bonding to their baby, which aligns with previous findings on partner support and emotional stability associating with maternal trajectories of increased mother–baby bonding (de Cock et al., 2016). Coparenting has also been previously associated with increased parental psychological flexibility (Yu & Xiao, 2021). About schooling, previous works addressing the postpartum period found, alike our worn findings, that higher educated mothers reported lower quality of mother–baby bonding (Bieleninik, Lutkiewicz, Cieślak, et al., 2021; de Cock et al., 2016). Mother–baby bonding of higher educated mothers may suffer from cross-over effects of stress in other areas of mothers’ life, namely increased workload, which has been found to predict parental stress (Tohme & Abi-Habib, 2022) that, in turn, negatively impacts on parent-baby bonding (Bieleninik, Lutkiewicz, Jurek, et al., 2021). Finally, there was a tendency for behavioural awareness to be higher in mothers who had more children, which contradicts previous findings using a measure of global parental psychological flexibility (Yu & Xiao, 2021). It may be that mothers of multiple children use their previous experience to be particularly more flexible and attentive to their owns’ and their babies’ behaviours, even if restricted (by having a newborn) of fully committing to valued actions.

This study has some noteworthy limitations that may limit the generalisation of results. The study had a cross-sectional design, involving a convenience sample of community mothers of babies up to 9 months postpartum. On the one hand that design does not allow us to explore how the relationship between variables may vary across time, as mothers become more adapted to their infant and their role as a parent; this should be further investigated using longitudinal designs. On the other hand, our results may thus be reflecting specific motherhood experiences and challenges from this postpartum period. For example, because we used an instrument initially designed to measure impaired bonding (i.e. PBQ), it may be less sensitive to variations of the quality of emotional bond in samples of low-risk mother–infant dyads, which is likely the case within the current sample. Moreover, because most of our sample (n = 208, 92%) was not receiving any psychological treatment, we could not explore the associations between psychological well-being, psychological flexibility and mother–infant bonding. As for our data analyses plan, we did not explore alternative models (e.g. the hypothesis that an experience of impaired mother–infant bonding might affect mothers’ perception of their attachment style or their psychological flexibility competencies rather than the other way around) nor non-recursive models. Our model was based on a priori theoretical assumptions that posited that prior more stable experiences (i.e. attachment representations) may impact on psychological and mutable processes (i.e. psychological flexibility), and then on how one copes with a new life experience (i.e. bonding with an infant). Exploring other (equally) sound theoretical models, particularly through longitudinal designs, may help to better understand the development of mother–infant bonding, and, subsequently, improve it when necessary. The final model adjusted for sociodemographic variables accounted for 29% of mother–infant bonding total variance, suggesting that other inter- and intrapersonal variables (e.g. baby temperament, social support, mother’s compassion abilities) might also be relevant. Still, and especially considering the scarcity of research addressing this topic, our findings represent a relevant contribution to further research on the modifiable variables referring to psychological processes and how they can contribute to the quality of mother–infant bonding. This is a topic of major importance, as the bonding quality has a significant impact on mother and infant well-being.

Overall, our findings highlight that the link between mothers’ attachment anxiety and difficulties in bonding with their child are at least partially explained by mothers’ being trapped in automatic pilot, thus having difficulty in being attuned with the present moment and in displaying behaviours that reflect their commitment to their own values. Conversely, a more secure orientation in relationships is associated with better bonding quality, only through mothers’ current abilities to be more aware and to freely choose and engage in meaningful behaviours that reflect their life values, instead of reacting unconnected to the present moment experience. Considering the modifiable nature of psychological flexibility, it is important to highlight the relevance and implications of our findings for future studies and intervention programmes aimed to promote more adaptive parenting practices and a more positive transition to motherhood. Evidence from studies with clinical populations has shown the efficacy of interventions enhancing psychological flexibility on improving parent–child interactions (e.g. Flujas-Contreras et al., 2021; Lobato et al., 2022). Therefore, our results suggest the relevance of including non-clinical, postpartum women from the community in preventive programmes focused on promoting psychological flexibility. This seems particularly relevant for those women suffering from attachment anxiety, and/ or for women that just had their first baby, are single and more educated. Our findigs seem to indicate that these are the women that may benefit more from additional support to ensure the quality of their bond with their babies.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by national funds through FCT – Fundação para a Ciência e a Tecnologia, I.P. (Portuguese Foundation for Science and Technology), within the project EXPL/PSI-GER/0259/2021.