Maternal sensitivity and children’s internalizing problems: the role of children’s gender and social skills Authors: Juanjuan Sun, Yuling Yin, Mengyu Xie, Pin Xu, Yan Li Published: 2026 ABSTRACT BackgroundThe development of internalizing problems in early childhood can have long-lasting negative impacts on social adjustment. Parental sensitivity is known to play a critical role in shaping children’s social skills and psychological well-being. However, the degree of this influence can vary based on the child’s gender, guided by the principles of gender role theory.ObjectiveThis study aims to explore the mediating role of children’s social skills and the moderating role of gender in the relationship between maternal sensitivity and internalizing problems in Chinese preschool children.MethodsA longitudinal design was used to investigate with 212 mother-child dyads in Shanghai, China. At Time 1 (T1, Mage =48.75), maternal sensitivity was assessed through mother-child interaction observation, and children’s internalizing problems was assessed through the mother’s self-report. At time 2 (T2, Mage = 60.28), mothers rate internalizing problems and teachers rate social skills.ResultsMaternal sensitivity at T1 significantly predicted lower internalizing problems at T2. Children’s social skills partially mediated the relationship between sensitivity and internalizing problems. Furthermore, child gender moderated the relationship between maternal sensitivity and social skills, with a stronger positive effect observed in girls.ConclusionThe findings highlight the protective role of maternal sensitivity against internalizing problems through enhanced social skills, particularly for girls. This underscores the importance of considering child gender in parenting interventions and contributes to a deeper understanding of the cultural context in early childhood development.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s40359-026-04670-8. Internalizing problems, encompassing anxiety and depression, represent a significant concern in child development. Unlike externalizing problems, which tend to decrease over time, internalizing problems exhibit an alarming upward trend throughout childhood and adolescence [1]. Recent studies indicate that over 20% of children experience clinically significant anxiety, with prevalence rates of anxiety disorders ranging from 5% to 17% in youth populations [2]. The ramifications of early internalizing problems are far-reaching, affecting classroom participation, interpersonal relationships, and even predicting delinquent behaviors in adolescence and adulthood [1, 3]. These findings underscore the critical need to elucidate the underlying mechanisms influencing internalizing problems in early childhood. While a substantial body of research has established links between parental sensitivity and children’s internalizing problems [4], meta-analyses reveal considerable heterogeneity in outcomes among children experiencing similar parenting behaviors [5, 6]. This variability suggests the presence of additional factors moderating this relationship. Previous research has pointed to children’s gender and social skills as potential sources of this variation, warranting further investigation [7]. The present study adopts an innovative approach by integrating gender role theory into the examination of maternal sensitivity and child internalizing problems. This theory posits that parents, particularly mothers, may exert a more profound influence on same-gender children due to shared socialization patterns and gender role expectations [8, 9]. This study seeks to explore these dynamics by examining how maternal sensitivity affects children’s internalizing problems through the mediation of social skills, while also considering the moderating role of the children’s gender. By exploring this dynamic, this study aims to shed light on the potentially differential impact of maternal sensitivity on boys and girls, a crucial consideration often overlooked in previous research. Parental sensitivity and children’s internalizing problems Bowlby-Ainsworth’s attachment theory underscores the critical role of maternal sensitivity in shaping children’s attachment security and long-term psychological development. According to the sensitivity hypothesis, a mother’s timely and sensitive responsiveness to her child’s needs fosters trust in her reliability and supportiveness, thereby facilitating the development of secure attachment [10, 11]. Secure attachment not only provides children with a “secure base” for exploration but also promotes positive cognitive, emotional, and behavioral outcomes. Highly sensitive parents provide a secure base from which children can confidently explore their environments [12], subsequently reducing the likelihood of internalizing problems such as withdrawal and anxiety [6, 13]. In contrast, children who experience insensitive, harsh, or rejecting caregiving often develop insecure attachment. This undermines their confidence in the mother as a source of emotional support and a foundation for safe exploration [14, 15]. These early caregiving experiences are internalized by children as a psychological representation known as the “internal working model” [16]. Research indicates that this model profoundly influences children’s expectations of caregivers and shapes their future cognitive, emotional, and behavioral development [17–19]. Thus, maternal sensitivity is not only central to the development of attachment security but also a key factor in children’s psychological well-being across cultural contexts. The protective role of parental sensitivity extends beyond biological relationships, as evidenced in studies of adoptive families [20]. Studies demonstrate that maternal sensitivity—the ability to respond to children’s signals in a child-centered manner—is a universal phenomenon, though its behavioral manifestations may vary across cultures [21]. This universal effect may be attributed to the enhanced self-regulation abilities of sensitive parents [22], enabling them to mitigate children’s negative emotions during stressful situations and offer timely comfort [23]. Conversely, children of less sensitive parents may develop negative self-schemas and avoidant coping strategies, potentially leading to internalizing problems [6, 24]. These children may perceive themselves as unworthy of love or their emotional experiences as inappropriate, resulting in the repression of negative emotions. In the Chinese cultural context, where the traditional ideology of “male dominates outside, female dominates inside” prevails, mothers often assume the role of primary caregivers. Consequently, maternal parenting behaviors emerge as stronger predictors of children’s problem behaviors compared to paternal influences [25]. However, most of the existing research on parental sensitivity relies on self-report measures, which may be subject to cultural biases and social desirability effects. Hendriks et al.’s [26] meta-analysis found that parent-reported and observed parenting styles share only 3% of variance, indicating limited consistency between these methods. This study addresses this gap by employing observational methods to assess maternal sensitivity in parent-child interactions within the Chinese cultural context. This approach not only enhances the ecological validity of our findings but also provides a more nuanced understanding of how culturally-specific parenting practices manifest in observable behaviors. By focusing on observed maternal sensitivity, we aim to capture the subtle nuances of mother-child interactions that may be particularly salient in Chinese families, thus offering a culturally-informed perspective on the relationship between maternal sensitivity and children’s internalizing problems. The mediating role of social skills between sensitivity and internalizing problems While the direct impact of maternal sensitivity on children’s internalizing problems is well-established, it is crucial to elucidate the mechanisms through which this influence occurs. As children enter kindergarten, their need for social interaction increases, highlighting the importance of social skills in establishing positive interpersonal relationships. Social skills, define as the ability to initiate social activities and display appropriate behaviors in interpersonal interactions [1], emerge as a potential mediator in this relationship. A growing body of research indicates that children with poor social skills struggle to establish and maintain interpersonal relationships, leading to negative emotions such as anxiety and loneliness, which in turn contribute to internalizing problems [27]. For instance, deficits in social skills during early childhood are associated with peer rejection and can predict psychopathological problems in late childhood and adolescence [1]. Attachment theory posits that early and ongoing experiences in parent-child relationships directly influence children’s social interaction patterns and play a crucial role in the development of social skills [28]. Specifically, maternal sensitivity has been identified as a significant predictor of children’s attachment patterns and social competence [24]. Highly sensitive mothers not only teach appropriate social skills but also provide positive role modeling of child-peer interactions, fostering the development of children’s social competence [6]. Children with well-developed social skills are more likely to initiate positive social interactions and gain peer acceptance, which may serve as a protective factor against internalizing problems [1]. On the contrary, children with poor social skills are unable to effectively respond to signals and needs of peers, leading to social pressure and anxiety, which in turn leads to internalizing problems Conversely, children with poor social skills may struggle to respond effectively to peer signals and needs, leading to social pressure and anxiety, ultimately contributing to internalizing problems [6]. Therefore, the present study hypothesizes that children’s social skills play a mediating role between maternal sensitivity and children’s internalizing problems. The moderating role of gender between sensitivity and social skills Child gender emerges as a critical factor in shaping social experiences, particularly in the context of parent-child interactions and peer relationships. Gender stereotypes and role expectations significantly influence parenting practices, with parents typically encouraging independence and exploration in boys while fostering caring, helpfulness, and close relationships in girls [29]. This differentiation in parental approach may have profound implications for the development of children’s social skills and, consequently, their internalizing problems. Research suggests that mothers tend to maintain closer relationships with their daughters, responding more promptly to their emotional needs and potentially enhancing their social skills to a greater extent. For instance, Fivush et al. found that mothers exhibit higher levels of sensitivity towards girls compared to boys [30]. This gender-differentiated sensitivity aligns with the theoretical model proposed by Bögels and Phares, which posits that child gender may moderate the relationship between parenting behaviors and anxiety outcomes [31]. Empirical evidence supports this moderation hypothesis. In an observational study, Krohne and Hock demonstrated that mothers of highly anxious girls were more likely to intrude on their children’s problem-solving behaviors compared to mothers of non-anxious girls, a relationship that did not hold for boys [32]. This finding underscores the potential for gender-specific effects in the link between maternal behavior and child outcomes. Social learning theory offers a compelling explanation for these gender differences. Parents are thought to take greater responsibility for the development of same-gender children and exert more influence on their behavior [31]. Moreover, girls tend to be more compliant and more readily internalize parental values compared to boys [9], potentially amplifying the impact of maternal sensitivity on their social skill development. Thus, child gender may play a moderating role between maternal sensitivity and children’s social skills. The present study Sensitivity, a key factor in forming different attachment patterns, directly influences children’s social skill and problem behaviors [6, 13]. Existing research has highlighted the significant impacts of maternal sensitivity on children’s internalizing problems. However, there are notable limitations that this study aims to address. Firstly, most research in this domain has employed cross-sectional designs, limiting the understanding of the long-term protective role of maternal sensitivity. This study adopts a longitudinal design, allowing for the exploration of these long-term effects. Secondly, there is a deficit in research focusing specifically on Chinese preschool populations, limiting the cross-cultural applicability of previous findings. This study addresses this gap by examining a sample of Chinese preschool children, providing valuable insights into how maternal sensitivity functions within a collectivist culture. Thirdly, few studies have explored the mechanisms behind the influence of maternal sensitivity on children’s internalizing problems, particularly from both the mother’s and children’s perspectives. This study innovatively investigates these mechanisms, incorporating theories of gender roles to examine how gender moderates the relationship between maternal sensitivity and children’s social skills (Fig. 1). Based on the foregoing, the present study proposes the following three hypotheses:Hypothesis 1 Maternal sensitivity will negatively predict children’s internalizing problems.Hypothesis 2 Children’s social skills will mediate the relationship between maternal sensitivity and internalizing problems.Hypothesis 3 Children’s gender will moderate the relationship between maternal sensitivity and children’s social skills, with maternal sensitivity having a greater impact on girls’ social skills. Hypothesis 1 Maternal sensitivity will negatively predict children’s internalizing problems. Hypothesis 1 Maternal sensitivity will negatively predict children’s internalizing problems. Hypothesis 2 Children’s social skills will mediate the relationship between maternal sensitivity and internalizing problems. Hypothesis 2 Children’s social skills will mediate the relationship between maternal sensitivity and internalizing problems. Hypothesis 3 Children’s gender will moderate the relationship between maternal sensitivity and children’s social skills, with maternal sensitivity having a greater impact on girls’ social skills. Hypothesis 3 Children’s gender will moderate the relationship between maternal sensitivity and children’s social skills, with maternal sensitivity having a greater impact on girls’ social skills. Fig. 1 Theoretical hypothesis model Theoretical hypothesis model Materials and methods Participants The initial sample of 212 children was recruited from two public kindergartens in Shanghai, China, using a random sampling approach. We specifically chose one kindergarten in an urban area and another in a suburban area, aiming to balance the sample while keeping the study feasible. Children with any diagnosed medical or psychological conditions were excluded to focus on typically developing preschool children. At Time 1 (T1), all 212 children (Mage = 48.75,52% boy) and their mothers participated in the study. One year later, at Time 2 (T2), 207 children (Mage = 60.28, 49% boy) and their mothers completed the follow-up assessment. The attrition rate was 4.52% (n = 9), primarily due to family relocation or withdrawal from the study for personal reasons. This low attrition rate suggests minimal bias due to participant dropout. Children’s mean age was 48.75 months (SD = 3.74). The sample included 108 boys (52%) and 99 girls (48%), with 166 (80%) being only children. Maternal demographic information indicated that 48 (23%) were full-time mothers, and 173(84%) had obtained a college degree or higher (see Table 1 for detailed sample characteristics). Table 1Basic family information (N = 207)Variable n %Child gender Boy10852.17 Girl9947.83Number of children Only one child15373.91 Two or more children5426.09Mothers’ working status Full-time mother4823.19 Working mothers15976.81Mothers’ educational degrees Senior school3416.42 Junior college4722.71 Bachelor’s degree9847.34 Graduate degree2813.53 Table 1Basic family information (N = 207)Variable n %Child gender Boy10852.17 Girl9947.83Number of children Only one child15373.91 Two or more children5426.09Mothers’ working status Full-time mother4823.19 Working mothers15976.81Mothers’ educational degrees Senior school3416.42 Junior college4722.71 Bachelor’s degree9847.34 Graduate degree2813.53 Basic family information (N = 207) The pattern of missing data across the measured variables, including child gender, child sibling status, maternal education, maternal sensitivity, child social skills, and child internalizing problems, was examined to ensure compliance with the assumption that the data were missing completely at random [33]. A Little’s MCAR test was conducted to rigorously evaluate whether the missing data could be considered random. The results showed that the assumption of MCAR was met (χ2 = 32.846, p = .107). This result provides statistical evidence that the missing data across these variables do not appear to be systematically related to the observed data or the missing values themselves, thereby reducing concerns about potential biases in the study. In addition, multiple imputation replaces missing data with possible values based on other available information in the dataset. Procedure This study was approved by the Ethics Committee of Shanghai Normal University. Written informed consent was obtained from all participants prior to data collection. The study employed a longitudinal design with two time points of data collection at age 4 and age 5, spaced one year apart. At T1, mothers and children participated in a structured parent-child interaction task, which was video-recorded for later coding. Mothers also completed the Children’s Social Competence and Behavior Evaluation scale. At T2, mothers completed the same scale again, while teachers assessed children’s social skills. Measures Sensitivity Maternal sensitivity was assessed at T1 using a structured observational task designed for early childhood interactions [34]. The task involved coding the first 10 min of mother-child interactions in 1-minute intervals with the Warmth and Control Coding System [35]. This system evaluates three dimensions—affec, sensitivity/responsivity, and control/guidance—assessing both positive and negative aspects of maternal behavior. Maternal positive sensitivity, defined as the mother’s ability to recognize and appropriately respond to the child’s needs, was scored on a 3-point scale: 0 (no response), 1 (moderate response, e.g., “Yes” or “You can put it there”), and 2 (high-level response, e.g., “You can put it there because we can use it for something else later”).Four independent coders, who underwent four months of training, conducted the coding [36]. The sensitivity scores were calculated using a time-sampling procedure, in which maternal sensitivity was rated every minute during the structured play task. The total ratings were summed and divided by the number of one-minute time-sample units to derive an average maternal sensitivity score. This methodology aligns with approaches used in prior longitudinal studies on early maternal sensitivity in Western contexts [37, 38]. To ensure coding accuracy, four independent coders underwent four months of rigorous training before conducting the ratings. Furthermore, to enhance the reliability of the coding process, four independent observers who were blinded to the study hypotheses double-coded 20% of the mother-child interaction videos. Any disagreements between coders were resolved through discussion until consensus was reached. Inter-rater reliability was calculated using two-way random-effects, single-measures ICC, with all four coders entered simultaneously into the model. The ICC for sensitivity was 0.88, indicating good inter-rater reliability. Social skills Children’s social skills were assessed at T2 using the Social Skills subscale of the Chines version of the Teacher Behavior Rating Scale [39, 40]. This 6-item subscale measures children’s social skills in peer interactions (e.g., inviting another child to play with him/her). Teachers rated each item on a 3-point scale (0 = Never, 2 = Always), with higher scores indicating better social skills. The social skills scores were calculated as the mean values. The scale demonstrated excellent internal consistency (Cronbach’s α = 0.89). Internalizing problems Children’s internalizing problem behaviors were measured at T1 and T2 using the Chines version of the Children’s Social Competence and Behavior Evaluation [41]. This 10-item subscale of “Anxiety-withdrawal” was used to examine children’s internalizing problems (e.g., “My child does not interact in groups”, 10 items). Mothers rated each item on a 6-point scale (1 = Never, 6 = Always), with higher scores indicating more severe internalizing problems. The internalizing problems scores were calculated as the mean values. The scale showed good internal consistency across both time points (Cronbach’s α = 0.87 at T1, 0.84 at T2, separately). Statistical analysis All analyses were performed using SPSS 26.0 and the PROCESS macro v4.0. Preliminary analyses included correlations and t-tests. We then tested for mediation using PROCESS Model 4 and moderated mediation using PROCESS Model 7, with 1,000 bootstrap samples for robust confidence intervals. All variables were standardized before analysis. To probe the significant interaction from the moderation analysis, we conducted a simple slope analysis. This standard procedure visualizes the nature of the interaction [42]. Specifically, because the moderating variable (child gender) has two levels, the relationship between the continuous independent variable (maternal sensitivity) and the outcome is plotted as two separate lines—one for boys and one for girls. To draw these lines, we plotted them using two representative points of maternal sensitivity: one standard deviation above (+ 1 SD) and below (-1 SD) the mean. This visualization method does not alter the continuous nature of the variable in the underlying statistical model. Results Preliminary analyses Preliminary analyses (Table 2) revealed significant correlations among key variables. Maternal sensitivity was positively correlated with children’s social skills and negatively correlated with internalizing problems. Children’s social skills were negatively associated with internalizing problems. However, no significant gender differences were observed in maternal sensitivity, children’s social skills, or internalizing problems. Table 2Means, standard deviations and correlation coefficients of major variablesMSD123451.T1Gender---2.T1Maternal sensitivity0.730.390.058-3.T2Social skills3.610.940.0370.182**-4.T1Internalizing problems2.220.58-0.048-0.159**-0.281**-5.T2Internalizing problems1.820.600.086-0.278**-0.320**0.250**-* p < .05, ** p < .01, *** p < .001 Table 2Means, standard deviations and correlation coefficients of major variablesMSD123451.T1Gender---2.T1Maternal sensitivity0.730.390.058-3.T2Social skills3.610.940.0370.182**-4.T1Internalizing problems2.220.58-0.048-0.159**-0.281**-5.T2Internalizing problems1.820.600.086-0.278**-0.320**0.250**-* p < .05, ** p < .01, *** p < .001 Means, standard deviations and correlation coefficients of major variables * p < .05, ** p < .01, *** p < .001 Analysis of the mediation effects To test the mediating role of social skills in the relationship between maternal sensitivity and children’s internalizing problems, we conducted a bootstrapped mediation analysis using PROCESS Model 4 with 1,000 resamples. All variables were standardized prior to analysis. We controlled for maternal education, maternal employment status, child sibling status, and T1 internalizing problems. Results (Table 3) indicated that maternal sensitivity at T1 significantly predicted children’s internalizing problems at T2 (β = -0.258, 95% CI= [-0.391, -0.124]). When social skills at T2 were included in the model, both sensitivity at T1 (β = -0.225, 95% CI= [-0.356, -0.094]) and social skills at T2 (β = -0.245, 95% CI = [-0.383, -0.107]) significantly predicted internalizing problems at T2. The indirect effect was significant (β = -0.031, 95% CI= [-0.070, -0.007]), suggesting that social skills partially mediated the relationship between maternal sensitivity and children’s internalizing problems. Table 3mediating effects of child social skillsPredictorsModel1(T2 Internalizing problems)Model2(T2 Social skills)Model 3(T2 Internalizing problems)βt95%CIβt95%CIβt95%CIT1Education0.1151.596[-0.027, 0.258]-0.058-0.775[-0.205, 0.089]0.1051.491[-0.034, 0.243]T1Sibling status0.1301.877[-0.007,0.266]0.0300.433[-0.108,0.169]0.1341.993*[0.001, 0.267]T1Working status-0.166-2.225*[-0.312, -0.019]0.0670.885[-0.083,0.217]-0.150-2.060*[-0.293, -0.006]T1Maternal sensitivity-0.258-3.807***[-0.391, -0.124]0.1612.383*[0.028, 0.294]-0.225-3.378**[-0.356, -0.094]T1Internalizing problems0.2043.080**[0.073,0.334]0.1432.153*[0.012, 0.275]T2 Social skills-0.245-3.496***[-0.383, -0.107] R 2 0.1660.0380.214 F 7.936***1.7879.021**** p < .05, ** p < .01, *** p < .001 Table 3mediating effects of child social skillsPredictorsModel1(T2 Internalizing problems)Model2(T2 Social skills)Model 3(T2 Internalizing problems)βt95%CIβt95%CIβt95%CIT1Education0.1151.596[-0.027, 0.258]-0.058-0.775[-0.205, 0.089]0.1051.491[-0.034, 0.243]T1Sibling status0.1301.877[-0.007,0.266]0.0300.433[-0.108,0.169]0.1341.993*[0.001, 0.267]T1Working status-0.166-2.225*[-0.312, -0.019]0.0670.885[-0.083,0.217]-0.150-2.060*[-0.293, -0.006]T1Maternal sensitivity-0.258-3.807***[-0.391, -0.124]0.1612.383*[0.028, 0.294]-0.225-3.378**[-0.356, -0.094]T1Internalizing problems0.2043.080**[0.073,0.334]0.1432.153*[0.012, 0.275]T2 Social skills-0.245-3.496***[-0.383, -0.107] R 2 0.1660.0380.214 F 7.936***1.7879.021**** p < .05, ** p < .01, *** p < .001 mediating effects of child social skills * p < .05, ** p < .01, *** p < .001 Analysis of the moderated mediation effects To examine whether child gender moderated the mediation pathway, we employed PROCESS Model 7 with 1,000 bootstrap resamples. As shown in Table 4, maternal sensitivity at T1 positively predicted children’s social skills at T2 (β = 0.152, 95% CI= [0.020, 0.283]). While child gender alone did not significantly predict social skills (β = 0.015, 95% CI= [-0.111, 0.141]), the interaction between maternal sensitivity and child gender was significant predicts social skills (β = 0.146, 95% CI = [0.117, 0.275]), indicating that child gender moderates the relationship between maternal sensitivity and children’s social skills. Table 4Testing the moderated effect of children genderPredictorsT2 Social skillsβt95% CIT1Gender0.0230.358[-1.05, 0.152]T1Education-0.043-0.589[-0.186,0.100]T1Sibling status-0.001-0.013[-0.137,0.135]T1Working status0.0470.633[-0.099,0.193]T1Maternal Sensitivity0.1932.889**[0.061,0.325]Sensitivity* Gender0.1892.886**[0.060,0.318] R 2 0.077 F 2.776** p < .05, ** p < .01, *** p < .001 Table 4Testing the moderated effect of children genderPredictorsT2 Social skillsβt95% CIT1Gender0.0230.358[-1.05, 0.152]T1Education-0.043-0.589[-0.186,0.100]T1Sibling status-0.001-0.013[-0.137,0.135]T1Working status0.0470.633[-0.099,0.193]T1Maternal Sensitivity0.1932.889**[0.061,0.325]Sensitivity* Gender0.1892.886**[0.060,0.318] R 2 0.077 F 2.776** p < .05, ** p < .01, *** p < .001 Testing the moderated effect of children gender * p < .05, ** p < .01, *** p < .001 To further explore how child gender moderates the relationship between maternal sensitivity and social skills, a simple effect was used to illustrate the moderating effect of child gender. As shown in Fig. 2, mothers’ sensitivity did not significantly predict boys’ social skills (β = 0.056, p > .05), whereas mothers’ sensitivity significantly predicted girls’ social skills (β = 0.330, p < .001). Fig. 2 Moderating role of child gender in the relationship between maternal sensitivity and child social skills Moderating role of child gender in the relationship between maternal sensitivity and child social skills Furthermore, as shown in Table 5, the moderated mediation analysis showed that child gender moderated the indirect pathway from maternal sensitivity to internalizing problems through social skills. The indirect effect was significant for girls (β = -0.073, 95% CI= [-0.161, -0.020]), but not for boys (β = -0.001, 95% CI = [-0.004, 0.052]). These results suggest that the mediating role of social skills in the relationship between maternal sensitivity and internalizing problems is more pronounced for girls than for boys. Table 5The mediating effect of social skills at different genderGenderEffectSE95% CIThe mediating role ofsocial skillsBoy-0.0010.026[-0.053, 0.052]Girl-0.1090.042[-0.209, -0.041] Table 5The mediating effect of social skills at different genderGenderEffectSE95% CIThe mediating role ofsocial skillsBoy-0.0010.026[-0.053, 0.052]Girl-0.1090.042[-0.209, -0.041] The mediating effect of social skills at different gender The mediating role of social skills Discussion The relationship between maternal sensitivity and child internalizing problems These longitudinal findings demonstrate that maternal sensitivity at Time 1 significantly predicted lower levels of children’s internalizing problems at Time 2, consistent with previous research highlighting the protective role of sensitive parenting against internalizing problems in early childhood [6, 13]. Firstly, sensitive mothers are adept at recognizing and responding appropriately to their children’s emotional needs [43]. This responsiveness facilitates children’s emotional regulation development, potentially minimizing the accumulation of negative affect that could otherwise contribute to internalizing problems. Sensitive mothers serve as a secure base, allowing children to effectively manage anxiety and distress [44]. Secondly, mothers exhibiting high sensitivity often possess strong self-regulation skills, which enable them to model effective emotional management strategies and offer timely comfort and support when children encounter stressors [28]. Through sustained positive interactions, sensitive mothers help children develop positive self-concepts and adaptive emotional expression strategies, which may buffer against the development of internalizing problems. Conversely, mothers lacking in sensitivity may fail to respond promptly or appropriately to their children’s emotional needs. This can lead to the formation of negative self-schemas in children, who may come to view themselves as unworthy of love or perceive their emotional experiences as inappropriate. Over time, this may foster the development of maladaptive coping strategies, such as emotional suppression, increasing the risk of internalizing problems [24]. It is worth noting that the protective effect of maternal sensitivity against internalizing problems may be particularly pronounced in the Chinese cultural context. Influenced by Confucian values, Chinese parents often view child-rearing as a fundamental responsibility, potentially leading to heightened sensitivity to their children’s needs [45]. This cultural emphasis on parental responsiveness may amplify the protective role of maternal sensitivity against children’s internalizing problems in Chinese families. The mediating role of children’s social skills The present study found maternal sensitivity indirectly reduces children’s internalizing problems by enhancing their social skills, thus identifying a crucial mechanism through which sensitive parenting influence children’s emotional well-being. This mediation effect aligns with previous research demonstrating the positive impact of maternal sensitivity on children’s social skills [6, 46]. According to attachment theory, children develop internal working models based on their early caregiving experiences, which subsequently guide their expectations and behaviors in future social interactions [12]. Sensitive maternal care fosters secure attachment, leading to the formation of positive internal working models of self and others. Consequently, children of sensitive mothers tend to approach social interactions with confidence and optimism, facilitating the development of advanced social skills. These enhanced social skills, in turn, serve as a protective factor against internalizing problems. Children with well-developed social skills are more likely to initiate and maintain positive peer relationships, providing them with a supportive social network that can buffer against emotional distress [1]. Moreover, successful social interactions reinforce children’s positive self-perceptions, further mitigating the risk of internalizing problems. Conversely, children exposed to insensitive parenting may develop insecure attachments, negatively impacting their beliefs and expectations about relationships [46]. These children perceive themselves as unworthy of care or view others as unreliable, hindering the development of effective social skills. As a result, they exhibit social withdrawal or experience difficulties in peer interactions, potentially leading to feelings of anxiety, loneliness, and other internalizing problems. The moderating role of children’s gender The present study further found that children’s gender moderated the relationship between maternal sensitivity and children’s social skills. Specifically, maternal sensitivity emerged as a stronger predictor of social skills in girls compared to boys. This gender-specific effect extends to the indirect pathway, with the mediation of social skills between maternal sensitivity and internalizing problems being more pronounced for girls. These results can be interpreted through the lenses of social role theory and social learning theory, offering insights into the complex interplay between parenting, gender socialization, and child development. Social role theory posits that gender stereotypes shape differential treatment of boys and girls, leading to gender-specific parenting behaviors [47]. In the context of social skills development, parents employ distinct strategies based on societal expectations of gender roles. For instance, parents often emphasize independence and exploration for boys, while prioritizing interpersonal skills and cooperation for girls, reflecting societal beliefs about traits critical for future success in gender-typical roles [29]. This differentiation results in mothers being more attuned to girls’ emotional cues and behaviors, fostering closer mother-daughter relationships. Indeed, research indicates that mothers more accurately predict their daughters’ distress, engage in more emotion-focused conversations with daughters, and maintain closer relationships with daughters compared to sons [30]. Social learning theory further elucidates this gender-specific effect by emphasizing the role of same-gender parents as primary models for children’s development [31]. Mothers feel a greater responsibility for socializing their daughters, potentially exerting more influence on girls’ social skill development. This theory suggests that mothers are more likely to use warm, supportive responses to model social skills for girls, with this modeling having a more significant long-term impact on daughters [47]. Moreover, girls tend to show greater conformity to parental behaviors and internalize parental values more readily than boys, potentially amplifying the effect of maternal sensitivity on girls’ social skills [48]. The gender-specific patterns in parent-child interactions may also contribute to these differences. For instance, mothers tend to be more attentive and responsive to girls’ negative emotions in stressful situations [49], while they employ harsher discipline or control in response to boys’ negative emotions or aggressive behaviors [9]. These differential responses reinforce gender-specific patterns in social skill development and, consequently, in the manifestation of internalizing problems. Limitations and contributions This study, while offering valuable insights, has certain limitations that warrant consideration. First, our sample was limited to young children and their mothers in mainland China. Given that parenting behaviors can vary across cultures, caution should be exercised when generalizing these findings to other cultural contexts. Future research should aim to replicate these results in diverse populations to establish their cross-cultural validity. Second, our focus was exclusively on maternal sensitivity. While this approach provided depth, it did not capture the potential influences of fathers and other caregivers on children’s internalizing problems. Future studies should expand the scope to include paternal sensitivity and explore potential differences between maternal and paternal influences on children’ s social adaptation. Notably, research could further investigate whether paternal sensitivity may relate more strongly to sons’ internalizing problems, as suggested by the moderating effects of child gender in maternal-child relationships (e.g., greater overlap between mothers and daughters). This avenue offers promising insights into the nuanced roles of caregivers in children’s psychological development. Despite these limitations, our study makes several significant contributions to the field. Theoretically, we integrate the analysis of maternal sensitivity, children’s social skills, internalizing problems, and child gender into a cohesive framework. This synthesis offers a comprehensive perspective on the factors associated with internalizing problems in early childhood. Moreover, by highlighting the moderating role of child gender, we contribute to the growing body of literature on how gender influences parenting practices and child outcomes. In addition, the use of observational measures of maternal sensitivity in parent-child interactions provides a more objective assessment compared to self-report measures [50]. This approach, relatively rare in Chinese contexts, offers valuable insights into the nuances of parent-child dynamics in this cultural setting. Furthermore, our longitudinal design allows for stronger inferences about the directional relationships between variables over time. The findings have significant practical implications for developing strategies to promote social adaptation in young children. By identifying social skills as a critical mediating mechanism, we provide a clear target for interventions. Education leaders and practitioners can use these insights to develop resources for parents to enhance their sensitivity and, consequently, improve their children’s social skills. The gender-specific findings underscore the need for tailored approaches, particularly in promoting boys’ social skills. Theoretically, these findings provide new evidence supporting social role theory in the context of parenting and child problems behavior. By demonstrating how children’s gender moderates the relationship between maternal sensitivity and social skills, it contributes to a more nuanced understanding of gender socialization processes in early childhood. Conclusions This longitudinal study reveals that observed maternal sensitivity significantly predicts lower levels of children’s internalizing problems one year later. This relationship is partially mediated by children’s social skills and moderated by child gender. Specifically, maternal sensitivity more strongly enhances girls’ social skills, indirectly reducing their internalizing problems. Supplementary Information Supplementary Material 1. Supplementary Material 1. Supplementary Material 2. Supplementary Material 2.