Research Review: Child emotion regulation mediates the association between family factors and internalizing symptoms in children and adolescents – a meta-analysis

Introduction

Internalizing disorders, typically characterized by depression and anxiety, are leading causes of disease burden globally and the most prevalent mental health disorders in children and adolescents (Polanczyk, Salum, Sugaya, Caye, & Rohde, 2015; Whiteford et al., 2013). Young people who experience elevated internalizing symptoms in childhood and adolescence are at greater risk of developing internalizing disorders in adulthood (Mulraney et al., 2021), which are associated with long-term health issues and functional impairments (Kessler et al., 2005). Understanding risk pathways that lead to elevated internalizing symptoms in young people is critical for improving targeted interventions for mental health problems (Mulraney et al., 2021). In particular, investigating key psychological and environmental risk factors that contribute to the development of internalizing symptoms, may be beneficial in this regard.

Theoretical models suggest that emotion regulation difficulties are a key underlying risk factor for internalizing symptoms in young people (Aldao, Gee, De Los Reyes, & Seager, 2016; Cole, Hall, Hajal, Beauchaine, & Hinshaw, 2017). Emotion regulation broadly refers to processes that influence the intensity and duration of emotions, as well as how they are experienced and expressed (Gross, 1998). Meta-analytic work in children and adolescents suggests that more frequent habitual use of adaptive emotion regulation strategies (cognitive reappraisal, problem-solving, and acceptance) is associated with fewer internalizing symptoms whereas the greater use of less adaptive strategies (avoidance, suppression, and rumination) is associated with greater symptoms (Schäfer, Naumann, Holmes, Tuschen-Caffier, & Samson, 2017). While most of the studies included in these meta-analyses were cross-sectional (limiting directional/causal inference), evidence from longitudinal studies does suggest that emotion dysregulation (e.g., rumination, deficits in emotional understanding, dysregulated emotional expression) prospectively predicts internalizing symptoms in young people (Cavicchioli, Tobia, & Ogliari, 2023; McLaughlin, Hatzenbuehler, Mennin, & Nolen-Hoeksema, 2011; Roelofs et al., 2009; Schneider, Arch, Landy, & Hankin, 2018).

The development of healthy emotion regulation is influenced by a complex interplay of innate and environmental factors, and research suggests that the family plays a crucial role in shaping the emotion regulation development of children and adolescents (Fosco & Grych, 2013; Ratliff et al., 2021; Repetti, Taylor, & Seeman, 2002). Within the field of developmental psychology, the tripartite model of child emotion regulation development is widely recognized (Morris, Silk, Steinberg, Myers, & Robinson, 2007). This model builds upon the theoretical foundations of social learning (Bandura & Walters, 1977), meta-emotion (Gottman, Katz, & Hooven, 1996), emotion socialization (Eisenberg, Cumberland, & Spinrad, 1998), attachment (Bowlby, 1988), and emotional security theories (Davies & Cummings, 1998), proposing three primary processes through which the family influences children's emotion regulation development, and in turn, internalizing and externalizing symptoms. The first process involves children's observation and modeling of how parents and others in the family respond to and regulate emotions. The second process involves parenting behaviors related to the socialization of emotions such as parents' reactions to children's emotions and discussions about emotions. The third process involves the emotional climate of the family, reflected in parenting styles, parent–child attachment, marital relations, parent–child relationships, and emotionality. Since it is difficult to isolate the effect of observation from emotion socialization behaviors and the emotional climate of the family (Morris et al., 2007), as well as the limited examination of observation/modeling as an individual factor in the literature, this meta-analysis will focus on the latter two components of the model.

Of all family factors, parent emotion socialization behaviors have been a focal point of interest in the literature. Parent emotion socialization behaviors are behaviors aimed at cultivating children's emotion regulation development and are broadly categorized into supportive and unsupportive (Eisenberg et al., 1998; Thompson & Meyer, 2007). Supportive emotion socialization behaviors include accepting, validating, and encouraging children's emotional expressions, empathizing and labeling emotions, as well as modeling healthy emotional expression and regulation. In contrast, unsupportive emotion socialization behaviors include punishing, denying, and dismissing children's emotions, or expressing distress and intolerance towards children's emotional expressions (Eisenberg et al., 1998; Fabes, Poulin, Eisenberg, & Madden-Derdich, 2002). There is robust evidence suggesting that both supportive and unsupportive parental emotion socialization are associated with emotion regulation abilities and internalizing symptoms in young people (see England-Mason & Gonzalez, 2020; Miller-Slough & Dunsmore, 2016 for reviews).

Other family factors are thought to influence child emotion regulation development via contributing to the emotional climate of the family (Morris et al., 2007). Longitudinal evidence shows that various parenting behaviors (parental warmth, responsiveness, harsh discipline, psychological control, behavioral control), parenting styles (authoritative and permissive styles), parent–child attachment style, and family conflict, are associated with emotion regulation abilities in children and adolescents (Boldt, Goffin, & Kochanska, 2020; Colman, Hardy, Albert, Raffaelli, & Crockett, 2006; Cui, Morris, Criss, Houltberg, & Silk, 2014; Horton et al., 2022; Jabeen, Anis-ul-Haque, & Riaz, 2013; von Suchodoletz, Trommsdorff, & Heikamp, 2011). Positive parent–child relationship, secure attachment style, and low levels of family conflict increase emotional security and encourage children to express emotions without the fear of judgment and rejection, which may help them to develop a healthy relationship with emotions and learn effective strategies to manage emotions in a safe and nurturing environment (Morris, Criss, Silk, & Houltberg, 2017).

The tripartite model also points to the importance of parent characteristics such as parents' own emotion regulation and psychopathology that can, directly and indirectly, impact child/adolescent emotion regulation. Empirical evidence has provided support for this notion. A recent meta-analytic review of 53 studies showed that adaptive parent emotion regulation was associated with higher positive parenting behaviors and adaptive emotion regulation in children (Zimmer-Gembeck, Rudolph, Kerin, & Bohadana-Brown, 2022). Mediation studies indicate that the link between parent and child emotion regulation may be mediated by family emotion expressiveness (Are & Shaffer, 2016) and emotion-related parenting behaviors (Morelen, Shaffer, & Suveg, 2016). In fact, parent emotion regulation has been identified as an intervention target to promote parental emotion socialization behaviors and emotion regulation abilities in young people (Hajal & Paley, 2020; Havighurst & Kehoe, 2017). Regarding parent psychopathology, Silk, Shaw, Skuban, Oland, and Kovacs (2006) found that children of depressed mothers were more likely to use ineffective emotion regulation strategies compared to children of non-depressed mothers in response to negative emotions induced in a delayed reward task. Parent internalizing psychopathology has also been found to be associated with child internalizing symptoms via its impact on child emotion regulation (Seddon, Abdel-Baki, Feige, & Thomassin, 2020).

In summary, the evidence linking family factors and child/adolescent emotion regulation coupled with the link between emotion regulation and internalizing symptoms supports the prominent tripartite model of child emotion regulation development, suggesting emotion regulation may mediate the relationship between family factors and internalizing symptoms in young people. However, this mediation model has not been systematically examined. Understanding this mechanistic pathway may provide new perspectives and directions for effective prevention and treatment of internalizing disorders in young people (Arango et al., 2018).

The present meta-analysis aimed to systematically investigate the mediating role of emotion regulation in linking various family factors and internalizing symptoms in children and adolescents. We hypothesized that positive family factors (e.g., supportive parent emotion socialization, positive family emotional climate, better parent emotion regulation and mental health) would be positively associated with better child and adolescent emotion regulation, which would, in turn, be negatively associated with child and adolescent internalizing symptoms. Given that parent emotion socialization behaviors have been theorized as particularly relevant in fostering emotion regulation development (Eisenberg et al., 1998; Thompson & Meyer, 2007), we hypothesized that supportive and unsupportive parent emotion socialization would provide the strongest evidence for mediation.

Methods

Results

Study characteristics

As shown in Figure 1, of 2,411 identified records, 49 studies reporting 138 analyses were included in meta-analyses. Of these, there were 15 longitudinal studies and 34 cross-sectional studies. There were 24,524 participants included in the study, and the mean age was 11.28 years. Most studies included a community-based sample (k = 46). Studies were conducted in 16 countries, with the majority of studies from the United States (k = 23). Study characteristics are summarized in Table S2. Based on prior literature (Pinquart, 2017; Yap, Pilkington, Ryan, & Jorm, 2014), a total of nine family factors were identified: supportive emotion socialization (k = 9), unsupportive emotion socialization (k = 10), psychological control (k = 7), behavioral control (k = 5), aversiveness (k = 8), secure attachment (k = 8), family conflict (k = 7), parent emotion regulation (k = 4), parent psychopathology (k = 4). Categorization was discussed at length and reached agreement by SCL, CK, EP, and SW. A summary of family factors is presented in Table 1.

Table 1. Summary of family factors with corresponding studies included in the meta-analyses

Family factor	Definition	Measure (and example item)	Study (in the meta-analyses)
Supportive emotion socialization	Accept, validate, and empathize children's emotional expressions, and positive modeling adaptive emotion regulation	Interview/vignette	Cunningham, Kliewer, and Garner (2009), Katz, Stettler, and Gurtovenko (2015), Moran, Root, Vizy, Wilson, and Gentzler (2019)
		CCNES (“when my child loses some prized possession and reacts with tears, I would tell him/her it's OK to cry when you feel unhappy”)	Han, Zhang, Davis, and Suveg (2020), Jin, Zhang, and Han (2017), Perry, Dollar, Calkins, Keane, and Shanahan (2020), Raval, Li, Deo, and Hu (2018)
		RAHAS (“if your child is very excited about going out with friends, I would tell him/her to have a good time”)	Raval, Luebbe, and Sathiyaseelan (2018)
		EAC (“When my child was sad, I took time to focus on her/him”)	Cui et al. (2020)
Unsupportive emotion socialization	Express distress and intolerance towards children's emotions, and punish, deny, and dismiss children's emotions	EAC (“when my child was sad, I told my child to stop being sad”) ESS (Portuguese version of EAC)	Azevedo, Ferreira, Martins, and Meira (2022), Buckholdt, Parra, and Jobe-Shields (2014), Cui et al. (2020)
		CCNES (“when I see my teenager becoming angry at a close friend, I usually tell him/her to not make such a big deal out of it”)	Han et al. (2020), Jin et al. (2017), Perry et al. (2020), Raval, Li, et al. (2018)
		RAHAS (“if your child is proud of him/herself for doing well on a test, I would tell him/her not to get so excited because one test does not make up one's whole grade”)	Raval, Luebbe, and Sathiyaseelan (2018)
		Observational task (PSI, EPI)	Yap, Allen, and Ladouceur (2008), Yap, Schwartz, Byrne, Simmons, and Allen (2010)
Psychological control	Parental behaviors to manipulate children's emotional experiences and expressions, such as the use of guilt induction, shaming, and conditional loving to pressure their children	PCS-YSR (“my mother/father is a person who is always trying to change how I feel or think about things”)	Cui et al. (2014), Luebbe (2014)
		DAPC (“My mother/father is disappointed when I do not depend on my mother/father for a problem”)	Ha and Jue (2018)
		CTQ emotional abuse subscale (”people in your family called him/her things like stupid, lazy, or ugly”)	Lin et al. (2019)
		CTS (“At home, my mother or my father makes me feel worthless”)	Padilla Paredes and Calvete (2014)
		CPRBI (“If I have hurt their feelings, stops talking to me until I please them again”)	Rueth, Otterpohl, and Wild (2017)
		FMSS (“we are just so close, I just do not want her to grow up”)	Han and Shaffer (2014)
Behavioral control	Parental monitoring of children's behaviors according to parents' expectations	Observational task during a pretend play and clean-up task	Perry, Dollar, Calkins, Keane, and Shanahan (2018)
		EPI task and PSI task (parents' direct commands and indirect commands were summed as the parental demandingness score)	Lo, Ng, and So (2021)
		PARQ/control (“my mother/father is always telling me how I should behave”) EMBU control (“your parent wants to decide how you should be dressed or how you should look”)	Di Giunta et al. (2022), Kallay and Cheie (2022)
		CRPBI (“my mother is a person who insists that I must do exactly as I am told”)	Rueth et al. (2017)
Secure attachment	Strong connections between parents and children, and low relationship anxiety and avoidance.	ECR-R avoidance subscale (“I prefer not to show how I feel deep down”) ECR-R anxiety subscale (“I worry about being abandoned”)	Brenning, Soenens, Braet, and Bosmans (2012), Brumariu, Kerns, and Seibert (2012), Shenaar-Golan, Yatzkar, and Yaffe (2021)
		ABR-Q (“My mother notices when I am feeling anxious”)	Iwanski, Lichtenstein, Mühling, and Zimmermann (2021)
		IPPA (“my mother respects my feelings”)	Chen, Wang, and Shao (2023), Chen, Zhang, Liu, Pan, and Sang (2019), Kullik and Petermann (2013)
Aversiveness	Parental hostility towards children, including criticism, punishment, harsh parenting practices, and emotional abuse	APQ subscale corporal punishment (“Your parents slap you when you have done something wrong”) CTS (“Hit or tried to hit with something”)	Balan, Dobrean, Roman, and Balazsi (2017), Lin et al. (2019)
		LEE (“My parents are sympathetic towards me when I'm ill or upset”)	Berla et al. (2022)
		FMSS (“he is a lazy person”)	Han and Shaffer (2014) Berla et al. (2022)
		HP (“he or she would use an object to beat the child when punishing them”)	Wang, Li, Wu, and Zhou (2022) Iselin et al. (2022)
		PARQ (“my mother/father punishes me severely when (s)he is angry”) EMBU (“you are the one whom your parent blames if anything happens at home”)	Di Giunta et al. (2022), Kallay and Cheie (2022)
Family conflict	inter-parental conflict, domestic violence, as well as parent–child conflict	CTS-2S/CTS (“destroyed something belonging to me or threatened to hit me”)	Harding, Morelen, Thomassin, Bradbury, and Shaffer (2013) Katz, Hessler, and Annest (2007)
		CPIC (“Even after my parents stop arguing they stay mad at each other”)	Lee et al. (2015)
		CEDV (hitting, punching, kicking, and shoving, calling names, swearing, yelling, threatening, and screaming)	Harding et al. (2013)
		MMEA (mothers reported on their own and their current partner's aggressive behaviors) CIPVI (30-min semi-structured interview focusing on specific episodes of inter-parental violence) CACI-2	Zarling et al. (2013)
		CBS (“my dad always seems to be complaining about me”)	Peng et al. (2021)
		FCS (“people in my family often insult or yell at each other”)	LaMontagne, Diehl, Doty, and Smith (2022)
		CSRI (“letting myself go, and saying things I do not really mean”)	Siffert and Schwarz (2011)
Parent emotion regulation	Parents' own emotion regulation ability	DERS (“when I'm upset, I become embarrassed for feeling that way”)	Crespo, Trentacosta, Aikins, and Wargo-Aikins (2017), Felton (2021), Seddon et al. (2020)
		RESE (“how well can you manage negative feelings when reprimanded by significant others?”)	Di Giunta et al. (2018)
Parent psychopathology	Parents' internalizing symptoms	CES-D (“I was bothered by things that usually do not bother me”)	Felton (2021)
		The Symptom Checklist-90-Revised (“how much were you bothered by feeling critical of others?”)	Suveg et al.(2011), Thomassin, Suveg, Davis, Lavner, and Beach (2017)
		BSI-18 (“how much were you feeling hopeless”)	Seddon et al. (2020)

• APQ, Alabama Parenting Questionnaire; BSI-18, Brief Symptoms Inventory-18; CACI-2, Computer-Assisted Child Interview – 2nd Edition; CES-D, Center for Epidemiological Studies-Depression Scale; CIPVI, Context of Intimate Partner Violence Interview; CPIC, Children's Perception of Interparental Conflict Scale; CSRI, Conflict Resolution Styles Inventory; CTQ, Childhood Trauma Questionnaire; CTS, Conflict Tactics Scales; CTS-2S, Revised Conflict Tactics Scale; DAPC, Domain-specific assessment of psychological control; DERS, Difficulties in Emotion Regulation Scale; EAC, Emotions as a Child Scales; ECR-R, Experiences in Close Relationships Scale - Revised; EMBU, Egna Minnen Betraffande Uppfostran; EPI, Event Planning Interaction; ERC, Emotion Regulation Checklist; ESS, Emotion Socialization Scale; FCS, Family Conflict Scale; FMSS, Five Minute Speech Sample; HP, Harsh Parenting; IPPA, Inventory of Parents and Peer Attachment; LEE, Levels of Expressed Emotion Scale; PARQ, Parent Acceptance-Rejection Questionnaire; PCS-YSR, Psychological Control Scale-Youth Self-Report; PSI, Problem Solving Interaction; RAHAS, Responses to Adolescent Happy Affect Scale; RESE, Regulative Emotional Self-Efficacy Scale.

Child and adolescent emotion regulation was most typically assessed by questionnaires such as Emotion Regulation Checklist (ERC; k = 8; Shields & Cicchetti, 1997), Emotion Regulation Questionnaires (ERQ; k = 5; Gross & John, 2003), and Difficulties in Emotion Regulation Scale (DERS; k = 4; Gratz & Roemer, 2004). Seven studies used observational measures or physiological indices of emotion regulation. Although emotion regulation was measured by a broad range of measures, we analyzed them under one construct as examining the mediating effect of different dimensions of emotion regulation was beyond the scope of our study. Studies assessed maladaptive emotion regulation strategies (k = 9) and emotion dysregulation (k = 17) were reverse scored to form a construct of child/adolescent emotion regulation in this meta-analysis. With regard to internalizing symptoms, most studies used the Centre for Epidemiologic Studies Depression scale (CESD; k = 13, Radloff, 1977), Child Behavior Checklist (CBCL; k = 10; Achenbach, 1999), and Youth Self Report (YSR; k = 6; Achenbach & Rescorla, 2001).

Mediation analyses

Table 2 shows the pooled correlations and 95% CIs for path a, path b, indirect effect, and total effect, together with study heterogeneity and mediation ratio. For direct effects, the paths from family factors to child/adolescent emotion regulation, and emotion regulation to internalizing symptoms were significant in all analyses. The direct path from all family factors (except behavioral control) to internalizing symptoms was significant. Heterogeneity of indirect effects was low to moderate. Forest plots of indirect effects are presented in Figure 2.

Table 2. Random effects of pooled correlation coefficients of path a, path b, indirect effect (a*b), total effect (path c), heterogeneity and mediation ratio

Family factor	k	n	Path a (95% CI)	I 2	τ 2	Path b (95% CI)	I 2	τ 2	Path a*b (95% CI)	I 2	τ 2	Path c (95% CI)	I 2	τ 2	[a*b/c]
Supportive ES	9	22	0.22 (0.15, 0.30)	34.73	0.00	−0.23 (−0.33, −0.13)	64.15	0.01	−0.05 (−0.10, 0.00)	0.00	0.00	−0.19 (−0.29, −0.08)	45.27	0.01	–
Unsupportive ES	10	18	−0.19 (−0.26, −0.12)	61.27	0.01	−0.34 (−0.42, −0.25)	86.32	0.04	0.06 (0.02, 0.11)	0.00	0.00	0.22 (0.13, 0.31)	71.07	0.01	0.27
Psychological control	7	16	−0.19 (−0.26, −0.12)	61.24	0.01	−0.30 (−0.37, −0.22)	75.30	0.01	0.07 (0.03, 0.10)	0.00	0.00	0.21 (0.11, 0.30)	79.86	0.01	0.33
Behavioral control	5	14	−0.18 (−0.23, −0.13)	0.00	0.00	−0.32 (−0.45, −0.18)	85.39	0.03	0.05 (−0.01, 0.10)	0.00	0.00	0.12 (−0.01, 0.24)	85.24	0.02	–
Secure attachment	8	26	0.26 (0.19, 0.33)	75.47	0.01	−0.31 (−0.40, −0.21)	89.46	0.02	−0.06 (−0.09, −0.03)	0.00	0.00	−0.33 (−0.42, −0.24)	88.16	0.02	0.18
Aversiveness	8	16	−0.25 (−0.32, −0.17)	69.64	0.01	−0.33 (−0.40, −0.25)	71.17	0.01	0.06 (0.03, 0.10)	0.00	0.00	0.26 (0.16, 0.35)	82.83	0.02	0.23
Family conflict	7	11	−0.28 (−0.36, −0.19)	81.32	0.01	−0.35 (−0.42, −0.26)	80.99	0.01	0.11 (0.08, 0.14)	8.63	0.00	0.42 (0.30, 0.53)	93.18	0.03	0.26
Parent ER	4	9	0.21 (0.10, 0.32)	82.17	0.01	−0.28 (−0.42, −0.12)	91.28	0.03	−0.07 (−0.13, −0.02)	29.76	0.00	−0.17 (−0.23, −0.11)	44.53	0.00	0.41
Parent psychopathology	4	6	−0.23 (−0.33, −0.13)	38.21	0.00	−0.31 (−0.37, −0.24)	6.53	0.00	0.08 (0.02, 0.14)	0.00	0.00	0.24 (0.03, 0.43)	84.07	0.04	0.33

• CI, confidence interval; ER, emotion regulation; ES, emotion socialization; I², I-squared statistics for heterogeneity; k, the number of studies, n, the number of analysis; τ², tau-squared statistic for heterogeneity.

Supportive emotion socialization

Nine studies with 22 analyses were included. Supportive emotion socialization was assessed by the meta-emotion philosophy interview and self-report measures, of which the most commonly used measure was The Coping with Children's Negative Emotions Scale (CCNES; k = 4, Fabes, Eisenberg, & Bernzweig, 1990). The pooled estimate of this indirect effect was not significant, p = .058.¹

Unsupportive emotion socialization

Ten studies with 18 analyses were included. Similar to supportive emotion socialization, unsupportive emotion socialization was most commonly assessed using CCNES. Two studies used observational parent–child interactions. The pooled estimate of this indirect effect was significant, p = .009. The percentage of the total effect explained by the indirect effect was 27%.

Psychological control

Five studies with 13 analyses were included. Psychological control was assessed by self-reported measures and observational tasks. The pooled estimate of this indirect effect was significant, p = .001. The percentage of the total effect explained by the indirect effect was 33%.

Behavioral control

Five studies with 14 analyses were included. Behavioral control was assessed by self-report measures and observational tasks. One study assessed autonomy support and was reversed scored. The pooled estimate of this indirect effect was not significant, p = .079.

Secure attachment

Eight studies with 26 analyses were included. Studies that assessed anxious, avoidant, and disorganized attachment were reverse scored and combined with studies that assessed secure attachment. The pooled estimate of this indirect effect was significant, p < .001. The percentage of the total effect explained by the indirect effect was 18%.

Aversiveness

Eight studies with 16 analyses were included. Aversiveness was assessed by self-report measures of corporal punishment, criticism, harsh parenting, or rejection. The pooled estimate of this indirect effect was significant, p = .001. The percentage of the total effect explained by the indirect effect was 23%.

Family conflict

Seven studies with 11 analyses were included. Family conflict was assessed by self-report measures of interparental conflict, parent–child conflict, or overall family conflict. The pooled estimate of this indirect effect was significant, p < .001. The percentage of the total effect explained by the indirect effect was 26%.

Parent emotion regulation

Four studies with nine analyses were included. Parent emotion regulation was mostly commonly assessed by DERS, and this self-report measure of emotion dysregulation was reverse scored. The pooled estimate of this indirect effect was significant, p = .012. The percentage of the total effect explained by the indirect effect was 41%.

Parent psychopathology

Four studies with six analyses were included. Parent psychopathology was assessed by self-report measures of parent depression or overall psychopathology. The pooled estimate of this indirect effect was significant, p = .014. The percentage of the total effect explained by the indirect effect was 33%.

Results of the separate meta-analyses for three developmental periods are reported in Table S4. Notably, more than 80% of the included studies had late childhood to early adolescent samples. When analyses were restricted to these studies, the results for most family factors were consistent with the main analysis. However, we found child emotion regulation mediated the relationship between supportive emotion socialization and internalizing symptoms, whereas parent emotion regulation did not support the mediation in the late childhood-early adolescent period. Due to the limited number of studies examining early-middle childhood and middle-late adolescence, meta-analysis was only feasible for a few family factors. We found significant mediation for family conflict in early-middle childhood, and no significant mediation for aversiveness and behavioral control in middle-late adolescence.

Discussion

This study examined the mediating role of emotion regulation in the relationship between family factors and internalizing symptoms in children and adolescents. We found evidence supporting mediation for some family factors, including unsupportive parental emotion socialization, psychological control, secure attachment, parental aversiveness, family conflict, parent emotion regulation, and parent psychopathology. However, we did not find mediation effects for supportive emotion socialization and behavioral control. Mediation effects were strongest for parent emotion regulation and psychopathology, followed by psychological control and unsupportive emotion socialization. In general, our results support the tripartite model of child emotion regulation development (Morris et al., 2007) and provide insights into specific aspects of family that may influence children's internalizing outcomes through emotion regulation.

Parental emotion socialization, an aspect of parenting that is thought to be particularly important for child emotion regulation development (Eisenberg et al., 1998; Thompson & Meyer, 2007), was hypothesized to have the strongest mediation effects. Partially supporting this hypothesis, we found that emotion regulation significantly mediated the relationship between unsupportive emotion socialization and child/adolescent internalizing symptoms. These findings align with prior research that links parental unsupportive emotion socialization behaviors (e.g., invalidating negative emotions) with emotion dysregulation and internalizing symptoms in young people (Kehoe, Havighurst, & Harley, 2014; Miller-Slough & Dunsmore, 2016; Shaffer, Suveg, Thomassin, & Bradbury, 2012). Supporting emotion socialization theory, these findings suggest that parental dismissing, invalidation, and punishment of children's emotions are likely to heighten emotional arousal and dysregulation in children (Eisenberg et al., 1998). Through such parenting behaviors, children may learn that negative emotions are inappropriate and should be avoided and minimized, which may undermine the healthy development of emotion regulation and lead to the development of internalizing problems.

However, we did not find a significant mediation effect of emotion regulation on the relationship between supportive emotion socialization and internalizing symptoms. The lack of mediation effect may be attributed to limitations in the current measures used to assess supportive emotion socialization. Specifically, the majority of the included studies used questionnaires (e.g., CCNES) that only measure parental encouragement of emotion expression and problem-solving reactions, while failing to capture the empathy and validation aspects of supportive emotion socialization. These measures are also unable to capture the temporal sequence of supportive emotion socialization behaviors. According to meta-emotion theory (Gottman et al., 1996), if parents respond to children's emotions immediately with problem-solving without validating or naming emotions first (which are fundamental steps of supportive emotion socialization), children may not experience emotion acceptance and therefore perceive parental responses as unsupportive (Kehoe, Havighurst, & Harley, 2020). Another possible explanation for the lack of the mediation effect is that the effect of supportive emotion socialization on children's emotion regulation and adjustment may be less pronounced than unsupportive emotion socialization, particularly for adolescents (Eisenberg et al., 1998). Adolescents, compared to younger children, are more likely to seek support from peers rather than parents (Laursen & Collins, 2009). Additionally, parents are less likely to respond to their adolescent children's emotions in a positive way due to increased rejection from adolescents (Hajal & Paley, 2020). Interestingly, an examination of developmental differences showed a significant mediation effect for supportive emotion socialization when considering only late-childhood to early-adolescent samples. Our meta-analysis lacked studies focusing on early-middle childhood, and only one study within the middle-late adolescence period reported a non-significant mediation effect. Therefore, the absence of a significant finding in middle-late adolescence likely influenced the overall non-significant result observed in our meta-analysis. These findings emphasize the crucial role of parent-supportive emotion socialization in fostering emotion regulation during the transition from childhood to adolescence, a period characterized by substantial development of emotion regulation and significant neurobiological maturation (Silvers, 2022).

Our findings suggest that psychological (but not behavioral) control is associated with child/adolescent internalizing symptoms indirectly via child emotion regulation. Psychological control, defined as parents' attempts to manipulate children's emotional and psychological expressions with guilt, shame, and conditional love, may interfere with children's development of emotional autonomy and, in turn, foster emotion suppression and non-acceptance. It may also limit children's opportunities to practice regulating emotions independently (Cui et al., 2014). Distinguished from psychological control, behavioral control involves parents' attempts to monitor children's behaviors with rules, guidelines, and monitoring (Barber, Stolz, Olsen, Collins, & Burchinal, 2005). One previous meta-analysis on parenting and child internalizing symptoms found that high levels of parental behavioral control were associated with low levels of symptoms, although behavioral control was a much weaker predictor of internalizing symptoms compared to psychological control (Pinquart, 2017). Contextualizing behavioral control within the broader framework of parenting styles (Baumrind, 1991), studies have found that authoritative parenting (which is characterized by high behavioral control and high warmth) is associated with better emotion regulation in children and adolescents while authoritarian parenting (high behavioral control and low warmth) is associated with poorer emotion regulation (Haslam, Poniman, Filus, Sumargi, & Boediman, 2020; Jabeen et al., 2013). Therefore, it is possible that behavioral control interacts with other dimensions of parenting (e.g., parental warmth) to influence children's emotion regulation development and internalizing outcomes.

Parent–child attachment, parental aversiveness, and family conflict, which contribute to the broader family emotional climate, were also implicated in significant mediation models. Secure parent–child attachment style is associated with parental acceptance of children's emotions and emotional responsiveness (Bowlby, 1988; Cassidy, 1994), which is critical for creating a safe environment where children can freely express their emotions. Previous research suggests that the attachment established in infancy can influence how parents respond to children's emotions throughout development, emphasizing the interconnected role of attachment and emotion-related parenting behaviors in shaping children's emotion regulation development (England-Mason & Gonzalez, 2020). On the other hand, parental hostility, criticism, punishment, and inter-parental conflict contribute to an emotionally unstable and unpredictable family environment, which undermines children's sense of security in expressing their emotional needs (Morris et al., 2007, 2017). The ability to express a wide range of positive and negative emotions is a key component of healthy emotion regulation, as it allows young people to receive support and learn to understand their emotional experiences, thus facilitating the development of effective regulatory strategies (Gross, 1998).

Particularly strong mediation effects were observed for parent characteristics (emotion regulation and psychopathology). As emotion dysregulation is recognized as a transdiagnostic feature of psychopathology (Aldao et al., 2016), it is possible that emotion dysregulation is the key phenotype driving parent psychopathology associations. Findings from this meta-analysis are consistent with extensive theoretical and empirical evidence supporting the intergenerational transmission of emotion dysregulation and psychopathology (Bariola, Gullone, & Hughes, 2011; Hajal & Paley, 2020), and our findings expand upon previous research by identifying child emotion regulation as a potential underlying mechanism. In addition to genetic influences, parents' own emotion regulation and psychopathology may impact children's emotion regulation development through its influence on modeling, parent emotion socialization practices, and the emotional climate in the family (Morris et al., 2007).

Conclusion

In conclusion, the meta-analyses conducted in this study provide evidence for the mediating role of emotion regulation in the relationship between various family factors and internalizing symptoms in children and adolescents. Our findings support the tripartite model of child emotion regulation development and highlight the importance of parent emotion regulation and mental health, emotion-related parenting behaviors, and family emotional climate in the development of children's emotion regulation and, in turn, internalizing symptoms. These findings have important implications for the development and refinement of parenting interventions to promote healthy emotion regulation development and better mental health in young people.