Maternal bonding in mothers of infants with excessive crying, sleeping, and/or eating disorders: the role of maternal stress and infant temperament

1 Introduction

Self-regulation represents a core developmental task in the earliest stages of life and refers to the ability to modulate arousal, emotion, attention and behavior (Rothbart et al., 2011). An imbalance between activation and inhibitory processes can lead to dysregulation and behavioral problems, manifesting as either excessive activation or inhibition (Papoušek et al., 2004). In infancy, intrapersonal regulatory capacities are still limited, so infants depend on the co-regulatory support of their caregivers. However, not all infants are equally successful in achieving this task. Infants with regulatory problems (RPs) have difficulties with self-regulation, often exhibiting excessive crying, difficult sleeping or eating problems. Within the broader construct of self-regulation, effortful control (EC) can be understood as a more specific temperamental component, describing the emerging capacity for the voluntary, top-down regulation of attention and behavior (Rothbart, 1991; Rothbart et al., 2011; Schmidt et al., 2025). Although related, effortful control and early RPs are conceptually distinct. Effortful control refers to a relatively stable individual disposition, whereas RPs describe observable, domain-specific difficulties in regulation. Importantly, low effortful control does not necessarily result in RPs, and early RP does not imply enduring deficits in effortful control. One way to further differentiate between low effortful control and early RPs may be to consider their temporal and contextual characteristics, with effortful control referring to a relatively stable temperamental trait, whereas early RPs are more likely to manifest in specific situations and may vary across contexts and developmental stages (Nigg, 2006; Rothbart et al., 2011; Sidor et al., 2017).

Excessive crying is defined according to the Wessel criterion (“rule of three”) as crying for at least 3 h per day on at least 3 days per week for a minimum of 3 weeks. Sleep onset disorder is defined by a sleep onset latency exceeding 30 min on most nights, whereas night waking disorder involves multiple or prolonged nighttime awakenings on most nights. Undereating disorder is characterized by reduced food intake relative to age, often accompanied by inappropriate eating behaviors such as lack of interest in food, selective eating, difficulties transitioning to solid foods, or eating during sleep (Zero to Three, 2016). RPs in infancy are linked with an increased risk of behavioral and mental disorders later in life (Asmussen et al., 2024; Galling et al., 2023; Hemmi et al., 2011; Wolke et al., 2023). Prevalence rates differ depending on the definition (problem vs. disorder), the population, the age group and the assessment method. In the first year of life, about 20–30% of infants show problems in one of the areas of crying, sleeping or eating (Cook et al., 2019; Hemmi et al., 2011; Petzoldt et al., 2016; Schmid et al., 2011). At the ages of 2–6 and 8–11 months, 2.9% and 8.6% of infants show problems in two or more areas, often referred to as multiple RPs (Olsen et al., 2019).

Theoretical models frame RPs within a relational context, including infant-related factors (e.g., difficulties in regulating emotions and behavior), parental psychosocial stressors (e.g., psychopathology or stress) and dysfunctional parent-infant interactions (Papoušek et al., 2004). A difficult temperament, defined by high negative affect (NEG), fussiness, and irritability is considered a correlate for RPs (Dale et al., 2011; Helseth et al., 2022; Martini et al., 2017; Mrakovits et al., 2025; Sidor et al., 2017). RPs are associated with increased parenting stress (Augustin et al., 2025; Bailhache et al., 2019; Schmid et al., 2011; Sidor et al., 2017), more social isolation and reduced social support (Lux et al., 2023; Martini et al., 2017; Papousek and von Hofacker, 1998), as well as increased depression and anxiety symptoms (Lux et al., 2023; Martini et al., 2017; Orton and Bilgin, 2026; Petzoldt et al., 2016), lower reflective functioning abilities (Georg et al., 2018), and differences in maternal mind-mindedness (Georg et al., 2025; Licata-Dandel et al., 2024). Mothers of infants with excessive crying, sleeping, and/or eating disorders tend to produce more appropriate and more non-attuned mind-related comments compared to controls. This increased verbalization of infants' mental states might stem from repeated attunement failures (Licata-Dandel et al., 2024).

Moreover, the quality of parent-infant interaction in parents with infants having RPs can be compromised (Olsen et al., 2019; Sidor et al., 2013). Mother–infant bonding constitutes a central context for the development of infants' self-regulation. Maternal bonding is a component of the reciprocal emotional bond between the primary caregiver and the child (Fallon et al., 2021; Klaus and Kennell, 1982). The bond begins during pregnancy, gradually strengthening during early childhood (Davies et al., 2021; Fallon et al., 2021). It refers to the degree to which the mother shows positive emotions, cognitive appraisals, and supportive actions toward her infant (Kinsey and Hupcey, 2013). Impaired bonding is reflected in a mother's feelings of irritability, hostility, or a sense of rejecting the infant (Hornstein et al., 2006). Figueiredo et al. (2009) describe bonding as a reciprocal process in which infants promote interaction through behaviors such as eye contact and crying. Disruptions in this bond are linked to more frequent infant RPs, including excessive crying (Lux et al., 2023; Sidor et al., 2013), higher colic ratings (Le Bas et al., 2020), and increased sleep problems (Hairston et al., 2016; Lux et al., 2023; Rusanen et al., 2025). Weaker mother–infant bonding is associated with poorer infant sleep, which in turn contributes to maternal insomnia and subsequently exacerbates bonding impairments (Kalmbach et al., 2022). However, Martini et al. (2017) report no significant association between maternal bonding and infant crying, sleeping, or feeding problems, expect for a marginally significant correlation with feeding problems. The fact that cues from infants with RPs are often unclear and hard to interpret can be highly demanding for parents. Repeated unsuccessful attempts by the mother to soothe their infant can result in significant physical and emotional exhaustion, generating negative feedback that may adversely affect the feelings toward the child (Garratt et al., 2019; Papoušek et al., 2004; Vik et al., 2009). Mothers may experience diminished parental efficacy, helplessness, and self-doubt regarding their parenting abilities, which can interfere with bonding processes (Cook et al., 2019; Papousek and von Hofacker, 1998). Furthermore, early RPs may foster strained interaction patterns, such as frequent misattunements (Licata-Dandel et al., 2024), potentially disrupting the development of a mutually positive mother-infant relationship and creating a vicious cycle of interactional deficits. Maternal bonding is considered a key determinant of infants' developmental trajectories and well-being. A meta-analysis suggests that stronger maternal bonding is correlated with easier temperament, greater attachment security, lower colic rating and enhanced positive affect. Moreover, higher bonding quality is linked to higher social-emotional scores, greater infant emotional stability, as well as lower social withdrawal (Le Bas et al., 2020, 2022). In turn, poorer maternal bonding is associated with a more difficult temperament (Davies et al., 2021; Doyle et al., 2023; Takács et al., 2020), and later internalizing and externalizing problems (Hatakka et al., 2025).

Maternal anxiety as well as depression symptoms are linked to higher bonding impairments (Davies et al., 2021; Doyle et al., 2023; La Rosa et al., 2025; Lux et al., 2023; O'Dea et al., 2023). Infant sleep problems are associated with impaired bonding only in the presence of maternal depression (Hairston et al., 2016). Lux et al. (2023) demonstrate that maternal bonding as well as social isolation mediate the associated between maternal psychopathology and infant sleeping problems. Other studies report an association between impaired bonding and lower social and partner support (Doyle et al., 2023; La Rosa et al., 2025), perceived stress (O'Dea et al., 2023) and poorer sleep quality (Kalmbach et al., 2022; La Rosa et al., 2025). Furthermore, mentalizing and parenting self-efficacy play important roles with regard to maternal–infant bonding (Qi et al., 2025).

To sum it up, several maternal and infant factors are associated with impaired bonding. Parents of infants with RPs often exhibit elevated maternal stress, higher levels of psychopathology, and reduced mentalizing abilities, all of which have been demonstrably linked to impaired bonding. While previous research has identified significant associations between these variables in mothers of infants with RPs, the present study extends this body of research in a clinical sample of infants with at least one diagnosis according to DC:0–5 (Zero to Three, 2016).

The present study aims to investigate whether maternal mentalizing abilities, parenting stress and psychopathology, as well as the degree of infants' RPs and infant temperament, are independently associated with maternal bonding in mothers of infants with excessive crying, sleeping, and/or eating disorders.

2 Material and methods

2.1 Study design and sample

The present study included a sample of N = 119 infants and their mothers. Participants were drawn as a sub-sample from an ongoing longitudinal project investigating psychosocial and (epi)genetic factors associated with early childhood disorders. The infants were either in inpatient or outpatient treatment at a socio-pediatric clinic in Southern Germany. Eligibility criteria for inclusion were: (1) infant age between 0 and 24 months, (2) sufficient maternal proficiency in German, and (3) a clinical diagnosis of one or more of the following disorders according to the DC:0–5 (Zero to Three, 2016): excessive crying, sleeping, and/or feeding disorders. Interested parents were referred to the study team. After receiving verbal information about the study and providing consent, participants were sent the study information, the declaration of consent as well as paper-pencil questionnaires. Completed questionnaire were returned postally to the study team. The study was approved by the Ethics Committee of the Technical University of Munich (vote number 34/19-S-SR) and was pre-registered (OSF, https://osf.io/xdv9y, registration date: 2025-01-09).

2.2 Measures

2.2.1 Bonding

Maternal bonding was assessed with the German version of the Postpartum Bonding Questionnaire (PBQ-16; Brockington et al., 2001; Reck et al., 2006). The questionnaire is based on 16 items screening bonding disorders. Answers were given on a 6-point Likert scale (0 = never agree to 5 = always). The total score was calculated by summing all items of the questionnaire, resulting in a possible score range from 0 to 80. High sum values indicate lower levels of bonding. The internal consistency has been proven to be good (Cronbach's α = 0.83).

2.2.2 Mentalizing abilities

Maternal mentalizing abilities were assessed with the German version of the Parental Reflective Functioning Questionnaire (PRFQ; Luyten et al., 2017). The questionnaire is based on 18 items covering the subscales “Pre-mentalizing” (PM), “Certainty about Mental States” (CM) and “Interest and Curiosity in Mental States” (IC). Answers were given on a 7-point Likert scale (1 = strongly disagree to 7 = strongly agree). Cronbach's Alphas in the present study are α = 0.48 (PM), α = 0.77 (CM) and α = 0.60 (IC).

2.2.3 Parenting stress

Parenting stress was measured using the German adaption of the Parenting Stress Index (PSI) (Abidin, 1995; Tröster, 2011). The questionnaire comprises 48 items that assess two domains: the parent domain, reflecting impairments in parental functioning, and the child domain, capturing stress related to the child's behavior. From the item responses, subscale scores as well as overall scores for each domain and a total stress score can be built. In this study, the overall parent domain score was used. The parent domain includes the following subscales: attachment, isolation, parental competence, depression, health, role restriction, and spouse-related stress. Answers were given on a 5-point Likert scale (1 = strongly agree to 5 = strongly disagree). Normative values differ between parents living with a partner and those without; therefore, specific T-scores were used to ensure comparability. According to the test manual, T-scores of 60 or higher indicate elevated parenting stress, whereas T-scores of 70 or higher indicate very high parenting stress (Tröster, 2011). The internal consistency of the parent domain in the present study has been proven to be good (Cronbach's α = 0.91). External validity has been proven in validation studies using different samples (Irlbauer-Müller et al., 2018).

2.2.4 Psychopathology

Maternal psychopathology was assessed with the Symptom-Checklist-90 (SCL-90-S; Franke, 2014). The questionnaire is based on 90 items covering psychological distress and symptoms of psychopathology. Answers were given on a 5-point Likert scale (0 = not at all to 4 = very strong). The SCL-90-S produces nine subscales (Somatization, Obsessive-Compulsive, Interpersonal Sensitivity, Depression, Anxiety, Anger-Hostility, Phobic Anxiety, Paranoid Ideation, and Psychoticism) and three Global Indices. In this study, the overall index of psychological symptom, the Global Severity Index (GSI), was used. The GSI was calculated by summing the responses across all 90 items and dividing the total score by the number of completed items. The resulting raw score was transformed into T-scores to ensure comparability. GSI T-scores below 60 are considered within the normal range, scores between 60 and 63 indicate a mild elevation, scores between 63 and 70 indicate clinically relevant psychological distress, and scores above 70 reflect marked psychological distress requiring further clinical evaluation (Franke, 2014). The GSI T-score threshold has been shown to be a valid and reliable indicator for identifying severe psychological distress. The internal consistency of the GSI has been proven to be good (Cronbach's α = 0.95).

2.2.5 Regulatory problems

Infants' RPs were assessed with the Questionnaire for Crying, Feeding and Sleeping (CFS; Groß et al., 2013). The questionnaire is based on 24 items for the subscale “Crying, Whining, and Sleeping”, 13 items for “Feeding” and 12 items for “Coregulation”. Answers were given on 4-point scale. Mean values were calculated (ranging from 1 to 4). Greater values indicate greater difficulties. In addition, an overall average value (Overall Regulatory Problems) can be built. The reliability of the questionnaire has been confirmed by the high internal consistency of the scales (Groß et al., 2013). Cronbach's Alphas in the present study are α = 0.79 (Overall regulatory problems), α = 0.88 (Crying, Whining and Sleeping problems), α = 0.91 (Feeding problems) and α = 0.75 (Coregulation).

2.2.6 Infant temperament

Infants' temperament was assessed with the Infant Behavior Questionnaire—revised very short form (IBQ-R VSF; Putnam et al., 2014) for infants under 18 months and with the Early Childhood Behavior Questionnaire—very short form (ECBQ VSF; Putnam et al., 2010) for infants of 18 months and older. The questionnaires are based on 37 items for the IBQ-R VSF and 36 Items for the ECBQ VSF covering the scales “negative affect” (NEG), “surgency” (SUR) and “effortful control” (EC). Answers were given on a 7-point scale (1 = never to 7 = always) with the eighth option, “Does not apply,” if that item is not relevant for their infant. The IBQ-R VSF has shown adequate internal consistency, test-retest reliability, and inter-rater agreement between mothers and fathers (Putnam et al., 2014). Composite scores were computed for negative affect, surgency and effortful control. Higher scores indicate higher levels of the respective temperament dimensions. Cronbach's Alphas for the IBQ-R VSF in the present study are α = 0.52 (SUR), α = 0.79 (NEG) and α = 0.77 (EC). Cronbach's Alphas for the ECBQ VSF in the present study are α = 0.95 (SUR), α = 0.77 (NEG) and α = 0.83 (EC).

2.2.7 Infant diagnoses

Diagnoses were made according on clinical criteria from the Diagnostic Classification of Mental Health and Developmental Disorders of Infancy and Early Childhood (DC:0–5; Zero to Three, 2016), the most specific classification system for diagnosing mental disorders in children aged 0–5 years. A second coding of the DC:0–5 diagnoses was conducted for 29.41% of cases (N = 35), yielding an interrater agreement of 94.74%. Any discrepancies were resolved through consensus discussion.

2.3 Statistical analyses

A power analysis was performed using G^*Power 3.1.9.7 (Faul et al., 2007, 2009) to determine the required sample size. Based on existing literature, the number of cases in the regression model described was estimated for low to medium effects. The calculation resulted in a total required sample size of N = 136 (α ≤ 0.05; 1 – ß ≥ 0.90).

To explore associations between bonding impairment and maternal and infant's factors, bivariate correlations were conducted. To detect if mentalizing abilities, parenting stress, maternal psychopathology, infant's RPs and infant's temperament were independently associated with bonding impairments, a multiple linear regression was conducted. The results were based on α = 0.05. Analyses were performed using IBM SPSS Statistics version 29.0.1 (IBM Corp., 2023).

3 Results

3.1 Descriptive analyses

Participant characteristics and descriptives can be found in Table 1. Sleep disorders were most prevalent, followed by feeding and crying disorders (Table 2). The descriptives of the study variables are depicted in Table 3.

Table 1. Sociodemographic characteristics of participants.

Table 2. Frequency of DC:0–5 diagnoses.

Multiple selection of different diagnoses possible.

Table 3. Descriptive statistics for study variables.

3.2 Inferential analyses

Associations of maternal and infant factors with maternal bonding are presented in Table 4. Correlational analyses indicated that greater bonding problems were associated with higher levels of maternal pre-mentalizing mode (r = 0.310, p < 0.001), lower certainty about mental states (r = −0.355, p < 0.001), parenting stress (r = 0.636, p < 0.001), and psychopathology (r = 0.527, p < 0.001). On the infant's side, greater bonding problems were related to increased crying and sleeping problems (r = 0.341, p < 0.001), lower levels of effortful control (r = −0.401, p < 0.001) and surgency (r = −0.205, p = 0.025). Interest and curiosity in mentals states (r = −0.179, p = 0.052) and negative affect (r = 0.178, p = 0.053) just slightly failed to reach significance with bonding (Table 4).

Table 4. Bivariate correlations between study variables.

^*p < 0.05. ^**p < 0.01, two-tailed significance level.

A multiple linear regression analysis including mentalizing abilities (pre-mentalizing mode, certainty about mental states), parenting stress, maternal psychopathology, infant crying and sleeping problems, and infants' temperamental traits (surgency, negative affect and effortful control) showed that maternal bonding impairment was significantly associated with maternal stress (B = 0.39, p < 0.001) and effortful control (B = −1.67, p = 0.016), while the other study variables were not significantly related, F_{(8, 118)} = 14.26, p < 0.001, Radj2 = 0.47 (Table 5).

Table 5. Results of linear regression examining maternal bonding problems.

3.3 Further analysis

In the mediation model, infant effortful control was significantly associated with maternal bonding impairment (B = −3.56, p < 0.001). Parenting stress was also associated with bonding problems (B = 0.56, p < 0.001). An indirect effect of effortful control on bonding problems via parenting stress was observed, indicating significant mediation (B = −1.40, 95% CI [−2.44; −0.05]). Despite the mediating effect of parenting stress, the direct effect of effortful control on bonding remained significant (B = −2.16, p = 0.001), indicating partial mediation (Figure 1).

4 Discussion

The aim of this study was to examine maternal bonding in a clinical sample of infants diagnosed with excessive crying, sleeping, and/or eating disorders. Based on correlational analyses, we found that maternal bonding impairments were associated with maternal variables such as pre-mentalization, certainty about mental states, parenting stress and psychopathology. On the infant's side, we found significant associations between maternal bonding problems and infant crying and sleeping problems as well as infant temperament, namely surgency and effortful control. In a multiple linear regression analysis, only parenting stress and lower infant effortful control showed significant independent associations with maternal bonding impairments. Interest and curiosity in mental states and negative affect fell just short of reaching statistical significance in their association with bonding problems. Further analyses indicated that the association between effortful control and bonding impairment was partially mediated by parenting stress.

Our finding that maternal bonding impairments were positively correlated with infants' crying and sleeping problems is in line with previous studies (Lux et al., 2023; Rusanen et al., 2025; Sidor et al., 2013), and highlights the relevance of bonding impairments in families affected by early RPs. In comparison, the descriptive means of maternal bonding reported by Reck et al. (2006) in a sample with depressive mothers were lower (M = 9.77, SD = 6.96) than those observed in our sample. This suggests that mothers of infants with excessive crying, sleeping, and/or eating disorders may be particularly vulnerable with regard to bonding difficulties because the infant's signals can be difficult to interpret. As a result, parents may struggle to identify what the infant needs and this can lead to feelings of helplessness, guilt and frustration (Garratt et al., 2019; Papoušek et al., 2004; Vik et al., 2009). Although a significant association with bonding problems was found for crying and sleeping problems, no significant associations emerged for overall regulatory problems, feeding problems, or coregulation. This pattern suggests that bonding is particularly relevant for domains closely linked to emotional and physiological regulation, which is consistent with previous research linking bonding problems to infant crying (Lux et al., 2023; Sidor et al., 2013) and sleeping problems (Hairston et al., 2016; Lux et al., 2023). In contrast, feeding problems may be more strongly influenced by physiological and routine-related factors (Kerzner et al., 2015). The lack of associations with the total score likely reflects the heterogeneity of the combined domains, which differ in their developmental relevance (Hemmi et al., 2011).

Likewise, maternal bonding problems were associated with perceived stress as reported before (O'Dea et al., 2023). Lux et al. (2023) reported associations between symptoms of depression and/or anxiety and bonding difficulties in a sample of infants with excessive crying and sleeping problems, which is in line with our findings. Mothers of infants with RPs often experience a high level of parenting stress (Augustin et al., 2025; Bailhache et al., 2019; Schmid et al., 2011; Sidor et al., 2013). High stress levels in parents are associated with a higher likelihood of maladaptive responses to excessive infant crying, which in extreme cases may include aggressive behaviors, such as shaking the infant, with serious consequences for the child (Barr et al., 2006; Zeifman and St. James-Roberts, 2017). This substantial strain can compromise sensitive caregiving and emotional availability, creating a stress-related cycle that may hinder the development of a healthy parent–child bond.

At the same time, our sample appears to be characterized by substantial resources, including low levels of psychopathology (GSI T-scores within the non-clinical range) and relatively few bonding impairments. In contrast, parenting stress was elevated, with mean PSI T-scores indicating a high level of parenting stress. Nevertheless, factors such as partnership stability as well as a relatively high level of education may contributed to this overall low-risk profile of the sample. PRFQ scores reflected relatively high levels of Certainty about Mental States and Interest and Curiosity in Mental States and were comparable to those reported in non-clinical samples (Georg et al., 2018). While significant associations between bonding problems and infants RPs were observed, these should be interpreted within the context of a sample that retains considerable protective resources. Accordingly, the findings do not reflect a high-risk population, but rather a group of mothers in which potential risk factors are likely buffered by multiple strengths, although the infants were highly burdened.

The observed association between bonding problems and maternal prementalizing, as well as certainty about mental states, can be interpreted in light of disruptions in reflective functioning. These difficulties may contribute to increased perceptions of bonding problems, as uncertainty or misinterpretation of infants' signals can interfere with the development of a secure and emotionally attuned parent-infant relationship. In this sense, impaired parental mentalization may represent a relevant cognitive-emotional pathway through which bonding difficulties emerge or are maintained in early parent-infant interactions. The observed association between pre-mentalizing modes and overall regulatory problems and crying and sleeping problems is in line with Georg et al. (2018), who argue that parents with a pre-mentalizing stance do not interpret infants' behavior as reflecting underlying mental states but rather as hostile or inexplicable. The present sample was characterized by relatively high levels of interest and curiosity regarding mental states, suggesting a group of mothers with a heightened attentional orientation toward mental states rather than consistently accurate and well-calibrated mentalizing. In this context, elevated interest and curiosity should not be equated with adaptive mentalizing per se, but may instead reflect increased vigilance or preoccupation with mental states in the absence of corresponding certainty or accuracy (see also Licata-Dandel et al., 2024). Accordingly, it appears plausible that interest and curiosity may be amplified to the point of hypermentalization, potentially taking on intrusive qualities; a similar pattern may also apply to certainty about mental states, which, when elevated without sufficient grounding in the child's actual experiences, may reflect overconfidence rather than genuine understanding (Luyten et al., 2017).

Results of the multiple linear regression analysis indicated that only parenting stress and infant effortful control showed significant associations with maternal bonding problems, suggesting that more proximal, interaction-related factors may be more directly related to bonding difficulties than broader maternal psychological characteristics (Abidin, 1992; Kinsey and Hupcey, 2013). Parenting stress likely reflects the immediate emotional and cognitive burden in caregiving, which may directly shape bonding experiences (Abidin, 1992). Similarly, infant effortful control may influence day-to-day interaction quality and transactional dynamics within the mother–infant relationship (Capelli et al., 2023; Takács et al., 2020).

Taken together, these findings support the conceptualization of the symptom triad (Papoušek et al., 2004) as a clinically meaningful cluster. This triad is understood within the context of interaction patterns, highlighting the association between infant-related factors (e.g., difficulties in regulating emotions and behavior), parental psychosocial stressors (e.g., psychopathology or stress) and dysfunctional parent-infant interactions (e.g., bonding). The present results highlight the association between maternal and infant variables as central factors shaping bonding experiences. Our findings provide valuable indications that infant RPs may contribute to increased parenting stress. At the same time, higher levels of parenting stress may reduce parental sensitivity an, which could increase infant's RP. Thus, the relationship between infant RP, parenting stress and bonding experiences may be understood as potentially bidirectional process. High levels of maternal stress may appear to exacerbate bonding difficulties, consistent with the notion that RPs challenge parental resources and increase the risk for negative emotional responses toward the infant. At the same time, infants with lower self-regulatory capacities may be more vulnerable to difficulties in the parent–infant relationship, thereby intensifying the impact of the symptom triad on bonding. Ambiguous or unpredictable infant signals, which are characteristic of infants with RPs (Papousek and von Hofacker, 1998), may increase parents' difficulties in accurately identifying and responding to their infants' mental states (Georg et al., 2025). Impaired bonding may negatively influence infants' regulatory capacities by limiting sensitive caregiving responses. This underlines that these dyads represent a risk group for negative developmental pathways. Considering the established association of early RPs with later infant socio-emotional difficulties and internalizing as well as externalizing problems (Asmussen et al., 2024; Galling et al., 2023; Hatakka et al., 2025; Hemmi et al., 2011; Wolke et al., 2023), the present results support the notion of a “double risk group” in the case of infants with RPs.

4.1 Strengths and limitations

Strengths of this study include the use of a well-characterized clinical sample and the application of the DC:0–5 (Zero to Three, 2016), which is considered the gold standard for diagnosing early crying, sleeping, and eating disorders. According to our knowledge, this is the first study investigating factors associated with maternal bonding in infants diagnosed with excessive crying, sleeping, and/or eating disorders according to DC:0–5 (Zero to Three, 2016). A post-hoc power analysis was conducted to evaluate statistical power. Based on the observed effect size (R² = 0.51, f² = 1.04) and the final sample size (N = 119), the analysis indicated very high statistical power (≈1.00).

However, several limitations need to be acknowledged. Our sample consisted predominantly of highly educated German-speaking mothers in partnerships which may limit generalizability. Previous research investigating regulatory problems has predominantly examined highly educated families in stable partnerships (Winsper and Wolke, 2014). To enhance generalizability, further research should include more diverse samples. The cross-sectional design precludes causal inferences and prevents examination of potential mediating mechanisms. Bidirectional effects—for example, that parenting stress also exacerbates RPs—cannot be ruled out. It is conceivable that parenting stress and effortful control affect maternal bonding but, in turn, that bonding impairment might negatively affect parenting stress and effortful control in terms of a vicious cycle. Future studies should use a longitudinal design to rule out the direction of effects. Moreover, the absence of a control group limits our ability to determine the extent to which bonding difficulties are specific to infants with RPs. Additionally, it must be mentioned that apart from the clinical rating of the DC:0–5 diagnoses, all other study variables were assessed by maternal self-report. The use of validated and widely accepted self-report instruments (e.g., PSI, SCL-90-S) improves methodological robustness and allows direct comparison with related studies. Still, results should be interpreted with some caution. It might be that mothers with higher levels of bonding problems perceive their infant's development more negatively, and vice versa. Further, it is possible that maternal reports of RPs and temperament may have been affected by underlying factors such as (psychological) distress, which may also influence the mother-infant bond. Maternal psychological distress is highly prevalent in the postpartum period and has been linked to maternal bonding (Kerstis et al., 2016; Moehler et al., 2006; Sasayama et al., 2025) and rating of infant temperament (Hanington et al., 2010). In addition, levels of maternal psychopathology were generally low in the sample (Table 3), with mean GSI T-scores indicating psychological distress within the non-clinical range. Although the sample was not characterized by elevated psychopathology, there was still sufficient variability in GSI T-scores to examine associations with the study. However, the relatively homogeneous and low-risk nature of the sample with regard to maternal psychopathology, bonding impairments, and parental reflective functioning may have limited the detection of stronger associations and restrict the generalizability of the findings to populations with higher levels of maternal psychopathology, parenting stress, or bonding impairments.

A limitation of the study concerns the relatively low internal consistency of the scale Pre-Mentalizing Mode of the PRFQ (Luyten et al., 2017) (α = 0.48). Although the exclusion of one problematic item led to a slight increase in reliability (α = 0.51), the item was retained to preserve the content validity of the scale, as it captures an important aspect of the construct that would otherwise not be represented. Therefore, the findings related to this measure should be interpreted with caution. At the same time, it should be noted that similarly low internal consistency coefficients for this measure (α = 0.54–0.59) have been reported in previous research (Georg et al., 2018; Kuchinke et al., 2026; Ramsauer et al., 2025; Wildner et al., 2024), suggesting that this issue may reflect broader challenges associated with the assessment of this construct rather than a limitation specific to the present study.

Furthermore, parental reflective functioning was assessed exclusively using an offline self-report questionnaire rather than during actual parent–child interactions. Self-report instruments such as the PRFQ assess parents' conscious evaluations of their own reflective capacities and may therefore be influenced by individual differences in self-awareness, insight, or socially desirable responding (Fonagy et al., 2016). In addition, questionnaire-based assessments may not adequately capture the dynamic and context-dependent nature of mentalization as it occurs during parent–child interactions. Furthermore, previous studies have reported inconsistencies in the factor structure and varying internal consistencies across different language adaptations of the PRFQ, further highlighting the need for cautious interpretation (Carlone et al., 2023; Moreira and Fonseca, 2023; Szabó et al., 2024; Wildner et al., 2024). Therefore, future studies should complement questionnaire-based assessments with observational or interviewer-based approaches to obtain a more comprehensive assessment of parental reflective functioning in specific interactional contexts.

Another limitation of the present study is that fathers were not investigated. Men also undergo a transition to fatherhood, with the postpartum period representing a particularly vulnerable phase (Koh et al., 2014). Paternal mental health and bonding during pregnancy and the postnatal period might be of equal critical importance, as they are closely lined with children's developmental outcomes (Ramchandani et al., 2013). The inclusion of fathers in future research would contribute to a more comprehensive understanding of determinants and outcomes of parental bonding.

4.2 Implications and future directions

Future studies should employ longitudinal designs to investigate the long-term consequences of early bonding impairments on infant psychopathology and socio-emotional development. Additionally, future research should consider discussing the inclusion of objective measures, such as additional objective assessments of infant temperament using the Laboratory Temperament Assessment Battery (LAB-TAB; Goldsmith and Rothbart, 1991; Planalp et al., 2017), as well as objective evaluations of parent–infant interactions through video-based observation, for example by assessing emotional availability (Biringen, 2022; Biringen et al., 2014). Future research, including fathers, may help to provide a more comprehensive understanding of the determinants and outcomes of parental bonding. Clinically, our findings underscore the vulnerability of mothers of infants with RPs to bonding difficulties. Early interventions—such as infant–parent psychotherapy—should address maternal bonding explicitly. Approaches that focus on enhancing parents' capacity for curiosity about their infants' inner world, such as mentalization-based group interventions (e.g., the Lighthouse program), may be particularly promising (Georg et al., 2021). Early support targeting parental sensitivity and stress management could strengthen bonding (Mascheroni and Ionio, 2019) and, in turn, reduce infants' RPs. A key implication of our findings for clinical practice is the importance of addressing maternal emotions and cognitions in the context of RPs. Interventions should openly acknowledge negative feelings that mothers may experience toward their infant, such as difficulties in bonding, and normalize such feelings as common among mothers of infants with RPs. Cognitive strategies that promote more realistic appraisals of bonding experiences can further support mothers.

Our findings are also highly relevant for psychoeducation. The causes and mechanisms underlying early RPs remain poorly understood and under-researched. Concerns about parental guilt regarding the origins of their infant's difficulties are particularly salient (Dobson et al., 2024). In line with this perspective, our findings showed that infant effortful control was significantly associated with RPs in the correlational analyses, suggesting that temperamental differences in infants' self-regulatory capacities may contribute to the expression of early RPs. Importantly, effortful control and RPs should be understood as conceptually distinct constructs: while effortful control reflects a temperamental capacity for voluntary regulation, RPs describe observable, domain-specific difficulties in regulation during infancy. Parents can be informed that early RPs may reflect an underlying biological disposition, such as limited effortful control, while also being shaped by parent-infant interaction processes and parental psychosocial stressors. Emphasizing this multifactorial perspective may alleviate parental burden, reduce guilt and shame in understanding RPs as challenges within the caregiving relationships rather as a result of parental failure. Ultimately, such interventions may prevent cascading risks for long-term socio-emotional difficulties.

Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Susanne Mudra, University Medical Center Hamburg-Eppendorf, Germany

Reviewed by: Christine Puckering, University of Glasgow, United Kingdom

Figen Eroglu Ada, Mimar Sinan Fine Arts University, Türkiye