Introduction
Adolescence marks an important period of increased risk for the development of psychiatric disorders, due to increased emotional sensitivity to interpersonal stressors (Casey et al., Reference Casey, Heller, Gee and Cohen2019; Dahl & Gunnar, Reference Dahl and Gunnar2009; Green et al., Reference Green, van de Groep, Sweijen, Becht, Buijzen, de Leeuw and Crone2021; Somerville et al., Reference Somerville, Jones and Casey2010). Past research has indicated that emotional reactivity is associated with adolescent psychopathology, including depressive symptoms, anxiety symptoms, and behavioral problems (Chiang et al., Reference Chiang, Chen and Chou2024a; Herres et al., Reference Herres, Ewing and Kobak2016; Kalvin et al., Reference Kalvin, Bierman and Gatzke-Kopp2016; Shapero et al., Reference Shapero, Abramson and Alloy2016). Family conflict generally predicts greater emotional reactivity. In adolescence, individuals also develop increased emotional sensitivity to specific types of stressors – namely peer stressors. Past research has shown that family conflict and hostile parenting are linked to increased youth sensitivity to peer rejection and problems (Godleski et al., Reference Godleski, Eiden, Kachadourian and Lucke2019; Guyer et al., Reference Guyer, Jarcho, Pérez-Edgar, Degnan, Pine, Fox and Nelson2015). However, limited research examined the direct association between family conflict and emotional reactivity to peer conflict. Moreover, it is unclear whether adolescents from high conflict families would exhibit greater emotional reactivity to peer conflict, which may subsequently predict the future development of internalizing psychopathology. To address these gaps in literature, this study investigated emotional reactivity to peer conflict as a mediator in the association between parent-adolescent conflict and adolescent internalizing problems.
Sensitization and adolescent psychopathology
The sensitization hypothesis posits that family conflict increases children’s reactivity to future conflicts in the family (Davies & Cummings, Reference Davies and Cummings1994). In support of the sensitization hypothesis, past research showed that, instead of habituating to conflicts, adolescents with a history of family conflicts tend to demonstrate heightened emotional reactivity to the later occurrence of conflicts (David & Murphy, Reference David and Murphy2004; Davies et al., Reference Davies, Thompson, Martin and Cummings2021; Goeke-Morey et al., Reference Goeke-Morey, Papp and Cummings2013; Pearson et al., Reference Pearson, Davies and Sturge-Apple2023). Emotion reactivity, defined as the extent to which an individual reacts emotionally to a stimulus (e.g., family conflict), is one dimension of emotion regulation. In past research, high levels of emotional reactivity have often been characterized by intense reactions to stimuli – reactions so intense that they are emotionally disproportionate or maladaptive (Agako et al., Reference Agako, Ballester, Stead, McCabe and Green2022; D’Agostino et al., Reference D’Agostino, Covanti, Rossi Monti and Starcevic2017; Gross & Feldman Barrett, Reference Gross and Feldman Barrett2011). Emotional security theory (EST) suggests that emotional reactivity reflects adolescents’ concerns about their security in the family (Davies et al., Reference Davies, Winter and Cicchetti2006) and such emotional reactivity is linked to the development of internalizing and externalizing problems (Buehler & Welsh, Reference Buehler and Welsh2009; Davies et al., Reference Davies, Cicchetti and Martin2012). Specifically, emotional reactivity serves as a mediator that bridges the association between family conflict and maladjustment (Davies et al., Reference Davies, Thompson, Martin and Cummings2021). In an attempt to preserve a sense of emotional security, children with high emotional reactivity are able to instantly identify potential threats of conflicts, and access psychophysiological resources to effectively adapt to them (Davies et al., Reference Davies, Martin, Sturge-Apple, Ripple and Cicchetti2016). However, such heightened status of arousal and vigilance would lead to prolonged and dysregulated distress (Davies et al., Reference Davies, Winter and Cicchetti2006). Thus, EST states that family conflict may undermine adolescents’ emotion regulation and that heightened emotional reactivity is one manifestation.
Much of the research on EST has focused on interparental conflict. Past research has shown that sensitization would also emerge from general family conflict and parent-adolescent interactions (Chiang et al., Reference Chiang, Chen and Liu2023, Reference Chiang, Chen and Chou2024a; Fosco & Grych, Reference Fosco and Grych2007; Herres et al., Reference Herres, Caporino, Cummings and Kendall2018). A history of family conflicts, either in the family in general or between parents and adolescents, would contribute to youth’s heightened emotional reactivity. For instance, youth from families with high levels of conflicts (e.g., we fight a lot in our family) reported lower emotional insecurity in the family (Cummings et al., Reference Cummings, Taylor, Merrilees, Goeke-Morey and Shirlow2016). In daily life, adolescents were angrier on days when they reported conflict with their parents (LoBraico et al., Reference LoBraico, Brinberg, Ram and Fosco2020). These findings suggest that adolescents may experience sensitization to parent-adolescent conflict, in addition to interparental conflict. Thus, adolescents who historically have a lot of conflict with their parents may be more reactive to future conflicts with their parents, which may then signal greater vulnerability to psychopathology. Building on these findings, the present study aims to extend the understanding of sensitization to conflict by investigating whether global parent-adolescent conflict is linked to adolescent emotional reactivity to conflicts in a different close relationship – peers – which in turn may be associated with adolescent psychopathology.
Family-of-origin sensitization to peer conflict
Family-of-origin sensitization refers to the concept that individuals exposed to destructive family conflict are more likely to view conflict in close relationships outside of family as a threat to their own emotional security, stimulating above-average emotional reactivity to conflict in peer or romantic relationships (Cook et al., Reference Cook, Buehler and Blair2013). That is, family conflicts may sensitize youth to conflicts in non-familial contexts, such as peer conflicts; heightened emotional reactivity to peer conflict can mediate the link between parent-adolescent conflict and subsequent psychological problems, especially in young adulthood. As a result, family serves as the original source of sensitization, exacerbating emotional reactivity in other relational contexts. Peer relationships are at the center of adolescents’ daily lives and well-being (Brown & Bakken, Reference Brown and Bakken2011; Chiang et al., Reference Chiang, Ting, Hung, Sung and Chen2024b; La Greca & Harrison, Reference La Greca and Harrison2005; Poulin & Chan, Reference Poulin and Chan2010). They are particularly important during the transition to college, when adolescents spend less time with family members and tend to live with peers and friends. Research has documented that family discords are associated with the quality of peer relations and friendships (Allen et al., Reference Allen, Hauser, O’Connor and Bell2002; Chung et al., Reference Chung, Flook and Fuligni2011; Schudlich et al., Reference Schudlich, Shamir and Cummings2004) but little research has documented the presence of family-of-origin sensitization and examined whether emotional reactivity to peer conflict is a mediating mechanism linking parent-adolescent conflict and adolescent psychological problems. Past research shows that exposure to high levels of conflict in one’s family of origin is detrimental to future relationships with peers. Adolescents exposed to interparental conflict perceived such conflicts as more threatening to their own well-being and the family, which increased the risk for social anxiety symptoms and subsequent lower levels of peer support (Weymouth et al., Reference Weymouth, Fosco, Mak, Mayfield, LoBraico and Feinberg2019). Similarly, in a 10-year longitudinal study, interparental conflict in childhood predicted increased emotional insecurity 5 years later in adolescence, and such insecurity predicted decreases in adolescent friendship affiliation (e.g., warmth, intimate disclosure) (Davies et al., Reference Davies, Martin and Cummings2018). The current study specifically tests family-of-origin sensitivity by testing the association between parent-adolescent conflict and emotional reactivity to peer conflicts in older adolescents.
Emotional reactivity to peer conflict may serve as a mediating mechanism between parent-adolescent conflict and adolescent psychopathology. The quality of early parent-child relationships shapes youth’s internal working models, which reflect their representations of both themselves and others. Insecure attachment styles lead to negative internal working models, which impede the development of future healthy relationships – an essential protective factor that promotes mental wellness (Herres et al., Reference Herres, Caporino, Cummings and Kendall2018; Kerns & Brumariu, Reference Kerns and Brumariu2014; Michiels et al., Reference Michiels, Grietens, Onghena and Kuppens2008; Verschueren et al., Reference Verschueren, Marcoen and Schoefs1996; Wang et al., Reference Wang, Wang, Wang and Wang2021). EST extends attachment theory and posits emotional reactivity as a developmental mechanism that links family conflict and psychopathology. Family conflicts lead to individuals becoming more sensitive and reactive to conflicts with peers, and viewing the conflict as threatening to their own and the relationship’s health (Cook et al., Reference Cook, Blair and Buehler2018). In turn, the frequent activation of strong negative emotions and the declining relationship quality may increase risk for internalizing problems. Cook et al. (Reference Cook, Buehler and Blair2013) found emotional reactivity in the context of interparental conflict was associated with reactivity to conflict in later friendship and romantic relationships. The findings suggest that family conflict may result in greater emotional reactivity to non-familial interpersonal stressors. Together, we argue that past family conflicts may put adolescents at greater risk for sensitization to peer conflict. This increased emotional reactivity, in turn, may contribute to poor mental health (Kleiman et al., Reference Kleiman, Ammerman, Look, Berman and McCloskey2014; Spear, Reference Spear2009).
Family-of-origin sensitization is developmentally appropriate to examine during the transition to young adulthood, when many youths are physically leaving their family of origin to live among peers. Relatedly, this transition is a key developmental milestone, and a time when youth gain significant levels of independence and form new relationships, often with same-aged peers. Whereas parents are primary partners in interpersonal emotion regulation during childhood and adolescence, peers take on this role during the transition to college, due to their greater physical. All the while, biological maturation underlying the development of emotion regulation capacity and autonomy from peer influences is still underway, well into mid-20s (Andrews et al., Reference Andrews, Ahmed and Blakemore2021; Sawyer et al., Reference Sawyer, Azzopardi, Wickremarathne and Patton2018).
Assessing emotional reactivity in daily life
One remaining gap is characterizing emotional reactivity in naturalistic settings using intensive longitudinal methods. Most research utilizes one-time questionnaires or laboratory observations to examine emotional reactivity to family or interpersonal discords. For instance, researchers used the well-validated scales, such as Emotional Reactivity Subscale of the Security in the Interparental Subsystem scale (SIS, Davies et al., Reference Davies, Forman, Rasi and Stevens2002), to evaluate adolescents’ emotional reactivity to interparental conflict (Buehler & Welsh, Reference Buehler and Welsh2009; Cook et al., Reference Cook, Buehler and Blair2013; Schermerhorn et al., Reference Schermerhorn, Cummings, DeCarlo and Davies2007). Other studies measured emotional reactivity elicited from structured interviews and tasks (Davies et al., Reference Davies, Cicchetti and Martin2012; Pearson et al., Reference Pearson, Davies and Sturge-Apple2023). Although these assessments provide robust evidence of one’s general level of emotional reactivity, intensive longitudinal methods such as daily diaries and ecological momentary assessment can further complement the existing literature by capturing emotional responses to specific types of stressful events as they naturally occur with reduced recall bias and high ecological validity (Bolger et al., Reference Bolger, Davis and Rafaeli2003; Laurenceau & Bolger, Reference Laurenceau and Bolger2005). Thus, the current investigation operationalizes emotional reactivity as the same-day association between peer conflicts and adolescent positive and negative emotions.
Additionally, limited research considers both positive and negative emotional reactivity. Most studies have focused on negative emotions (e.g., sad, fearful, angry) (Cook et al., Reference Cook, Blair and Buehler2018; Davies et al., Reference Davies, Cicchetti and Martin2012; Kelley et al., Reference Kelley, White, Milletich, Hollis, Haislip, Heidt and Henson2016). Nevertheless, positive emotional reactivity may also provide unique information on adolescents’ emotional responses that are largely unknown in literature. Positive emotions have unique contributions to the development of adaptive capacity and personal resources, including cognitive (e.g., enhancing learning performance), physical (e.g., health conditions, psychological (e.g., resilience), and social (maintaining relationships) (Gilbert, Reference Gilbert2012; Taylor et al., Reference Taylor, Rosenfield, Dowd, Dutcher, Hofmann, Otto, Pollack and Smits2023). Positive emotional reactivity may represent a strong indicator of emotion dysregulation with significant declines in positive emotions in the face of distress and negative events.
The present study
The present study examined the association between parent-adolescent conflict, emotional reactivity to peer conflict, and adolescent mental health problems, and tested the mediating role of emotional reactivity to peer conflict in the link between parent-adolescent conflict and mental health problems. This study extended prior research by using a daily diary approach with a 6-month follow-up assessment. Rather than measuring emotional reactivity using a single administration questionnaire, this study captured daily emotional reactivity by examining within-person relations between peer conflict and emotions. In addition, our Taiwanese sample differs from the past predominantly North American and European samples. Given the cultural emphasis on family relations in collectivistic societies, testing family-of-origin sensitization in this sample may enrich our understanding of the EST and sensitization hypothesis. As shown in Figure 1, we had three hypotheses regarding the family-of-origin sensitization. First, we aimed to evaluate the associations between parent-adolescent conflict and emotional reactivity to peer conflict, including positive and negative emotional reactivity. We hypothesized that higher parent-adolescent conflict would be associated with greater positive and negative emotional reactivity to peer conflict, testing the hypothesis of family-of-origin sensitization. Second, we examined whether emotional reactivity to peer conflict was prospectively associated with mental health problems as indicated by depressive symptoms and anxiety symptoms. We hypothesized that greater emotional reactivity would be associated with more depressive symptoms and anxiety symptoms. Next, we evaluated whether emotional reactivity to peer conflict would mediate the prospective relationships between parent-adolescent conflict and mental health problems.
Figure 1. Conceptual model of family-of-origin sensitization and mental health problems.
Method
Participants
Participants consisted of 108 Taiwanese first-year college students between ages 18 and 19 (M age = 18.53, SD age = 0.39; 64% female) recruited in Taiwan. All participants were born in Taiwan and the majority (96%) of their parents were Taiwanese (3% Chinese, 1% Southeast Asian). To be eligible for participation, participants needed to meet the following inclusion criteria: (a) access to mobile devices most of time to complete online surveys; (b) living with parents most of time before entering college; (c) contact with parents at least once a month (35% currently living with parents). Participants reported parent education level ranging from below high school degree (6%), high school degree (23%), college degree (59%), and graduate degree (12%), as well as family income ranging from low-income (4%), middle-to-low (17%), middle (63%), upper-to-middle (14%), and rich (2%). Baseline and daily diary surveys were conducted between December 2022 and January 2023, and the follow-up assessments were conducted in June 2023.
All study procedures were approved by the affiliated University’s Institutional Review Board. Participants were recruited by referral and snowball sampling from three universities in north and central Taiwan. The research team reached out to college faculty who invited eligible students to enroll in the study via classroom announcements. The research team explained research goals and procedures to interested participants. After informed consent, participants completed a general questionnaire about demographic information, family and peer relationship, health behaviors, and well-being. Starting from Monday, participants completed 14-day ecological momentary assessment four times per day. Given the focus on peer conflict and emotion, this study utilized evening surveys, which asked about peer conflict and positive and negative emotions. Each night, participants received daily surveys through emails or apps with person-specific links at 7 pm. The compliance with daily surveys were excellent (99.5% of total prompts completed). Participants were compensated with $35 USD for their participation. Six months later, the participants completed the follow-up assessments and completed questions about their mental health. All participants completed baseline and follow-up assessments.
Measures
Parent-adolescent conflict
Adolescents rated the extent to which they had conflicts with parents on three items adapted from the Parent-Adolescent Relationship Scale in the Taiwan Youth Project, a large-scale, longitudinal panel study in Taiwan (Yi et al., Reference Yi, Wu, Chang and Chang2009): “I argued with my parent a lot” “My parent and I were angry toward each other,” and “I had conflicts with my parent.” Past studies have shown that the parent-adolescent conflicts subscale has good validity and reliability (Cheng & Wang, Reference Cheng and Wang2021; Chiang & Bai, Reference Chiang and Bai2022; Jou et al., Reference Jou, Wu and Huang2010). Items were rated on a five-point scale ranging from 1 (strongly disagree) to 5 (strongly agree). Internal reliability (Cronbach’s α) was 0.81.
Depressive symptoms
The Patient Health Questionnaire-9 (PHQ-9; Kroenke et al., Reference Kroenke, Spitzer and Williams2001) was used to assess depressive symptoms at the baseline and follow-up surveys. PHQ-9 has excellent psychometric validity and reliability for assessing depressive symptoms among clinical and community samples (Fonseca-Pedrero et al., Reference Fonseca-Pedrero, Díez-Gómez, Pérez-Albéniz, Al-Halabí, Lucas-Molina and Debbané2023; Gao & Liu, Reference Gao and Liu2024; Richardson et al., Reference Richardson, McCauley, Grossman, McCarty, Richards, Russo and Katon2010). Adolescents were asked to report how often they have experienced depressive symptoms during the past two weeks with nine items such as “little interest or pleasure in doing things,” “feeling down, depressed, or hopeless,” and “trouble falling or staying asleep, or sleeping too much.” Items were rated on a four-point scale ranging from 0 (not at all) to 3 (nearly every day) and averaged. Internal reliability (Cronbach’s α) was 0.86 at baseline and 0.85 at follow-up. Higher scores indicated greater depressive symptoms.
Anxiety symptoms
The Generalized Anxiety Disorder-7 (GAD-7; Spitzer et al., Reference Spitzer, Kroenke, Williams and Löwe2006) assessed the severity of anxiety symptoms. Research has shown that GAD-7 has great psychometric properties (e.g., specificity, sensitivity) for detecting the severity of youth anxiety symptoms (Mossman et al., Reference Mossman, Luft, Schroeder, Varney, Fleck, Barzman and Strawn2017; Tiirikainen et al., Reference Tiirikainen, Haravuori, Ranta, Kaltiala-Heino and Marttunen2019). Adolescents were asked to report how often they have experienced anxiety symptoms during the past two weeks on seven items such as “feeling nervous, anxious, or on edge,” “worrying too much about different things,” and “trouble relaxing.” Items were rated on a four-point scale ranging from 0 (not at all) to 3 (nearly every day) and averaged. Internal reliability (Cronbach’s α) was 0.84 at baseline and 0.87 at follow-up. Higher scores indicated greater anxiety symptoms.
Daily peer conflict
Adolescents completed two items assessing peer conflict modified from past daily diary studies (Chiang et al., Reference Chiang, Ting, Hung, Sung and Chen2024b): “I argued with my friend(s) today” and “I did not get along with my friend(s) today.” Adolescents were asked to rate how much they experienced peer conflict during the day on a five-point scale from 1 (not at all) to 5 (a lot). Given that peer conflict was assessed daily, frequency can be inferred from a summative score (e.g., number of days when the individual selected >2 on the Likert scale). Nonetheless, the focus of the current study was on same-day responses to peer conflict, and the Likert scale yielded greater variability on the day-level predictor. Reliability was good for within-person (R C = .80) and between-person (R 1F = .89) levels. Items were averaged to create a single score of peer conflict.
Daily positive and negative emotion
Adolescents reported their daily emotions on ten items adapted from the Positive and Negative Affect Schedule-Short Form (PANAS-SF, Thompson, Reference Thompson2007). We assessed positive emotion with five items (i.e., determined, active, attentive, inspired, alert) and negative emotion with five items (i.e., nervous, sad, hostile, ashamed, afraid). Adolescents were asked to rate their emotion during the day on a five-point scale from 1 (not at all) to 5 (a lot). Reliability was good for within-person (positive: R C = .89; negative: R C = .82) and between-person (positive: R 1F = .94; negative: R 1F = .90) levels. Items were averaged to create daily scores of positive and negative emotions.
Analytical plan
This study examined the association between parent-adolescent conflict, emotional reactivity to peer conflict, and psychopathology indicated by depressive and anxiety symptoms. To test the hypotheses, we conducted two-step analytical procedures using R. In the first step, we captured daily emotional reactivity to peer conflict using two Multilevel Models (MLM): one estimating negative emotions and the second estimating positive emotions. Consistent with past research (Mroczek et al., Reference Mroczek, Stawski, Turiano, Chan, Almeida, Neupert and Spiro III2015; Ong et al., Reference Ong, Gardner, Urganci, Gunaydin and Selcuk2020; Santee & Starr, Reference Santee and Starr2022), we calculated emotional reactivity in MLM by deriving the within-person slopes of the daily associations between person-mean-centered peer conflict and positive and negative emotions. Specifically, each person has within-person random intercepts and slopes wherein the intercepts represent the average levels of emotions, and the slopes indicate the unique association between peer conflict and emotions on a given day. These slopes reflect the degree to which changes in individual’s peer conflicts predict the corresponding changes in positive and negative emotions across the daily diary period, suggesting each individual’s emotional reactivity to peer conflict. Based on these, we derived Empirical Bayes estimates of each individual’s slope (Chiang & Bai, Reference Chiang and Bai2024; Santee & Starr, Reference Santee and Starr2022), which quantified each adolescents’ unique relationship between peer conflict and positive and negative emotions as an index of emotional reactivity to peer conflict. Thus, these estimates documented the levels of emotional reactivity for each person (i.e., between-person index). For positive emotional reactivity only, we recoded the scores (multiplying by −1) so that a higher score indicates that the individual experienced greater declines in positive emotions on days when they experienced more peer problems than was usual for them (Bai & Repetti, Reference Bai and Repetti2018). Higher emotional reactivity refers to greater increases in Negative Emotion (NE reactivity) or greater declines in Positive Emotion (PE reactivity).
In the second step, we included the indices of NE and PE reactivity in a single Structural Equation Model (SEM) to test their mediation effect in the relationship between parent-adolescent conflict and psychopathology. SEM was conducted using lavaan package in R (Rosseel, Reference Rosseel2012) with full information maximum likelihood method. We included a latent variable of parent-adolescent conflict (indicated by the three items) and NE and PE reactivity scores in a SEM model with depressive symptoms and anxiety symptoms as the dependent variables. Residual errors of positive emotional reactivity and negative emotional reactivity were allowed to correlate. Given that Chi-square (χ2) is vulnerable to sample size, we evaluated the model fit using the following fit indexes: Root Mean Square Error of Approximation (RMSEA), Comparative Fit Index (CFI), and the Standardized Root Mean Square Residual (SRMR). The model fit was considered acceptable when RMSEA was below 0.08, CFI was above 0.95, and SRMR was below 0.05 (Bollen, Reference Bollen1989; Hu and Bentler, Reference Hu and Bentler1999). To evaluate the mediation effect of emotional reactivity, we utilized a bootstrap procedure to test the indirect effect (MacKinnon et al., Reference MacKinnon, Lockwood, Hoffman, West and Sheets2002). Following the recommendations (Mallinckrodt et al., Reference Mallinckrodt, Abraham, Wei and Russell2006; Shrout & Bolger, Reference Shrout and Bolger2002), we used 95% bias-corrected confidence intervals with 5,000 bootstrapping resamples to calculate the mediation effect in the analyses.
Results
Descriptive statistics and correlations were presented in Table 1. The within-day correlations between peer conflict and positive and negative emotions were −0.10 (p < .001) and 0.34 (p < .001), respectively. The results of MLM analyses showed that, for the average adolescent, more peer conflicts were associated with lower positive emotions (b = −0.37, p < .001) and higher negative emotions (b = 0.49, p < .001) on that day. Further there was significant individual variability in PE and NE reactivity to peer conflicts, indicated by significant likelihood ratio test of variance components (p < .001). We then extracted these slopes to represent the unique association between peer conflict and emotions for each adolescent in SEM models. To test our main hypotheses, we specified one structural model estimating pathways between global parent-adolescent conflict, PE and NE reactivity to peer conflict, and adolescent depressive symptoms and anxiety symptoms. We conservatively included baseline levels of depressive symptoms and anxiety symptoms.
Table 1. Descriptive statistics and correlations
* p < .05, **p < .01,***p < .001.
Given recent work suggesting that a significant association between the predictor and the outcome is not a prerequisite for mediational analyses (Hayes, Reference Hayes2009; O’Rourke & MacKinnon, Reference O’Rourke and MacKinnon2018), we tested the integrated model in which emotional reactivity to peer problems mediated the association between global parent-adolescent conflict and depressive symptoms. The mediational model fit the data well, χ2 = 89.45, p < .01, RMSEA = 0.04, CFI = 0.96, SRMR = 0.04. As shown in Figure 2, higher level of global parent-adolescent conflict was associated with greater NE reactivity to peer conflict (b= .25, p < .05) but not associated with PE reactivity to peer conflict (b= .11, p > .05). Next, greater NE reactivity to peer conflict was associated with more depressive symptoms (b= .57, p < .001) and anxiety symptoms (b= .50, p < .001). PE reactivity to peer conflict was also associated with depressive symptoms (b= .23, p < .05) but not associated with anxiety symptoms (b= .12, p > .05). To examine the potential mediation effect of NE reactivity, we used bootstrapping method with 5,000 resamples to calculate the 95% confidence intervals around the indirect effect. The indirect effect from global parent-adolescent conflict to depression symptoms via NE reactivity to peer conflict was significant (b= .12, p < .05; 95% CI [0.04, 0.22]), indicating that NE reactivity mediated the prospective association between global parent-adolescent conflict and depressive symptoms. To test whether NE reactivity fully or partially mediated the association, we conducted a constrained model that evaluated changes in model fit when limiting the path from global parent-adolescent conflict to depressive symptoms as zero. The constrained model did not significantly differ from the original model, Δχ2(1) = 1.42, p > .05, suggesting that NE reactivity full mediated the association between global parent-adolescent conflict and depressive symptoms.
Figure 2. Structural equation modeling of the association between global parent-adolescent conflict, emotional reactivity, and depressive symptoms and anxiety symptoms. Paths were standardized coefficients. The model controlled for outcomes at baseline. *p < .05, **p < .01, ***p < .001.
Next, the results showed that the indirect effect of global parent-adolescent conflict on anxiety symptoms via NE reactivity to peer conflict was significant (b = .13, p < .05; 95% CI [0.03, 0.19]), indicating that NE reactivity mediated the prospective association between global parent-adolescent conflict and anxiety symptoms. We constrained the path from global parent-adolescent conflict to anxiety symptoms as zero to test whether NE reactivity partially or fully mediated the prospective association. However, the constrained model significantly differed from the original model, Δχ2(1) = 4.20, p < .05, suggesting that NE reactivity only partially mediated the association between global parent-adolescent conflict and anxiety symptoms.
Discussion
Guided by the sensitization hypothesis (Cummings & Davies, Reference Cummings and Davies2002; Goeke-Morey et al., Reference Goeke-Morey, Papp and Cummings2013), this study extends previous findings by examining the mediating effect of emotional reactivity to peer conflict on the prospective associations between global parent-adolescent conflict and adolescent psychopathology. The study uniquely assessed both PE and NE emotional reactivity using a sample of adolescents who reported the daily levels of peer conflict and positive and negative emotions. Findings showed that higher global parent-adolescent conflict was associated with greater NE reactivity to peer conflict, which in turn was associated with more depressive symptoms and anxiety symptoms 6 months later. Mediation analyses showed that NE reactivity to peer conflict mediated the prospective association of global parent-adolescent conflict with depressive symptoms and anxiety symptoms. Moreover, PE reactivity to peer conflict was associated with more depressive symptoms. Our results revealed that sensitization to interpersonal conflicts may originate from parent-adolescent conflict and influence adolescents’ negative emotional reactivity to conflicts outside the family.
The findings extend past studies that primarily focused on sensitization to specific types of conflict – particularly, interparental conflict (e.g., Davies et al., Reference Davies, Cicchetti and Martin2012; Goeke-Morey et al., Reference Goeke-Morey, Papp and Cummings2013). We found that youth who experienced more conflict with their parents reported greater negative emotions in response to peer conflict, which in turn predicted higher levels of depression and anxiety symptoms. Youth with greater parent conflict may perceive peer conflicts as more threatening or indicative of poor relational quality, leading to heightened negative emotions (Cook et al., Reference Cook, Buehler and Blair2013). It is also possible that repeated exposure to family conflict – a form of chronic stress – undermines the youth’s ability to react and recover adaptively in the face of future stressors, such as peer conflict. Whether through an increased sense of poor relationship quality or dysregulation of the affective stress response system, parent-child conflict increases the risk of depression and anxiety (Byrd et al., Reference Byrd, Vine, Beeney, Scott, Jennings and Stepp2022; Kretschmer et al., Reference Kretschmer, Sentse, Meeus, Verhulst, Veenstra and Oldehinkel2016; Michiels et al., Reference Michiels, Grietens, Onghena and Kuppens2008). This pattern aligns with previous research showing that greater NE reactivity predicts the future development of psychopathology (Davies et al., Reference Davies, Cicchetti and Martin2012; Santee & Starr, Reference Santee and Starr2022; Stepp et al., Reference Stepp, Scott, Jones, Whalen and Hipwell2016). Furthermore, NE reactivity to peer conflict mediated the prospective associations between global parent-adolescent conflict and depressive and anxiety symptoms. These findings support the sensitization hypothesis and emotional security theory, which propose that family conflict exacerbates children’s emotional reactivity, thereby increasing their vulnerability to psychological problems (Cummings & Davies, Reference Cummings and Davies2002; Goeke-Morey et al., Reference Goeke-Morey, Papp and Cummings2013; Pearson et al., Reference Pearson, Davies and Sturge-Apple2023). The results suggest that sensitization originating within the family can heighten adolescents’ NE reactivity in non-familial contexts, supporting the family-of-origin sensitization hypothesis. This study is the first to find that parent-adolescent conflict was associated with heightened emotional reactivity to peer conflict, which then predicted future risk of depressive symptoms and anxiety symptoms among adolescents. Thus, conflict with parents appears to undermine emotional security in new relationships that adolescents form throughout young adulthood, further increasing risk for psychopathology.
Moreover, the findings suggest that greater PE reactivity to peer conflict (i.e., more decreases in positive emotions) was prospectively associated with increases in depressive symptoms, offering new insights into the importance of positive emotions. Positive emotions support adaptive responses to distress and promote coping and regulation in stressful situations (Taylor et al., Reference Taylor, Rosenfield, Dowd, Dutcher, Hofmann, Otto, Pollack and Smits2023). It is noteworthy that the evaluation of positive emotional responses adds to the existing literature, which has predominantly focused on negative emotions. The majority of studies have concentrated on NE reactivity, often overlooking the potential impact of PE reactivity on psychological well-being (e.g., Davies et al., Reference Davies, Forman, Rasi and Stevens2002; Eisenberg et al., Reference Eisenberg, Fabes, Nyman, Bernzweig and Pinuelas1994; Nock et al., Reference Nock, Wedig, Holmberg and Hooley2008). A meta-analysis across 58 published papers exploring the relationship between child maltreatment and the development of emotion reactivity revealed a notable absence of investigations into experiences of positive emotions (Lavi et al., Reference Lavi, Katz, Ozer and Gross2019). This gap in the literature underscores the limited exploration of PE reactivity, with research in its early stages (Chiang et al., Reference Chiang, Chen and Chou2024a). Consequently, our findings contribute to the literature on emotional reactivity by finding that PE reactivity to peer conflict predicts the emergence of depressive symptoms, thereby advancing our understanding of the nuanced interplay between positive emotions and psychological outcomes.
Our finding that NE reactivity plays a more prominent role in family-of-origin sensitization is aligned with prior work on emotion dynamics and psychological well-being. Emotion dynamics, defined as patterns, changes and fluctuations in emotional states over multiple points in time, is associated with individual differences in their psychological well-being; this association is much stronger for negative emotions than positive emotions, suggesting that negative emotion dynamics constitute a more significant target of clinical interventions (Houben et al., Reference Houben, Van Den Noortgate and Kuppens2015). Consistent with the emotion security theory, parent-child conflict in the family of origin likely sensitizes individuals to experience more intense and overwhelming negative emotions when their emotion security is threatened (Buehler & Welsh, Reference Buehler and Welsh2009; Davies et al., Reference Davies, Cicchetti and Martin2012). We found that PE reactivity predicted more depressive symptoms, consistent with the tripartite model of anxiety and depression, wherein distress (i.e., negative emotion) characterizes both anxiety and depression and the absence of positive emotion differentiates depression from anxiety (Clark & Watson, Reference Clark and Watson1991).
Study findings have translational implications. Specifically, they point to several distal and proximal targets interventions for the prevention of depression and anxiety during the transition to college. First, as documented in prior research, high levels of conflict between parent and adolescents pose risk for more depression and anxiety symptoms during the transition to young adulthood. Existing programs to foster healthy parent-adolescent relationships (Van Ryzin et al., Reference Van Ryzin, Roseth, Fosco, Lee and Chen2016), are important for youth mental health. One mechanism by which parent-adolescent conflict contributes to depression and anxiety symptoms is via increased negative emotion reactivity to peer conflict in non-familial contexts. Thus, a more proximal prevention interventions for the reduction of depression and anxiety symptoms in young adulthood may include healthy relationship development with an emphasis on peers and romantic partners, and negative emotion regulation.
This study had several methodological strengths. By using the daily diary design, we assessed adolescents’ emotional reactivity through daily reports of peer conflicts and emotions, which reduced recall bias and provided a more objective evaluation of emotional reactivity. Next, the longitudinal design allowed us to examine the prospective association over six months while controlling for baseline levels of the outcomes. Third, the high compliance rate across baseline, daily diary, and follow-up assessments in the study can minimize potential attrition biases because of participant attrition and missing data.
Despite these strengths, several limitations of this study should be noted and addressed in future research. First, the results may not be generalizable to other families with diverse racial/ethnic backgrounds. However, this study represents one of the few studies that investigated sensitization hypothesis in non-Western samples, and we found evidence in support of the sensitization hypothesis in Taiwanese adolescents. Future studies could expand on the sensitization hypothesis by investigating cultural influences on the pathways between family conflict, emotional reactivity, and adolescent psychopathology. Second, our sample was not recruited from a clinical population, potentially constraining the applicability of the findings to youths with greater vulnerability to psychiatric illness. Nonetheless, these findings may have relevance for adolescents at early risk of psychopathology during the transition to college when life stressors increase and mental health deteriorates (Fruehwirth et al., Reference Fruehwirth, Mazzolenis, Pepper and Perreira2023). Third, it is important to acknowledge that the study relied on adolescent reports to test our hypotheses. Future research would benefit from applying multi-informant designs to examine family conflicts and adolescent emotional reactivity from diverse perspectives. Fourth, we only focused on the level of peer conflicts, which may not adequately capture the full scope of peer conflicts such as the frequency, intensity, and resolutions of peer conflicts. Examining various aspects of peer conflicts is an important next step for future research, as investigating the differences in peer conflicts will further clarify whether adolescents would exhibit different levels of emotional reactivity. Fifth, emotional reactivity was defined as maladaptive responses to stressors in the current study, which assumed a linear association between emotional reactivity and mental health problemsFootnote 1 . However, a mild level of emotional reactivity to significant stressors may be adaptive and normative reactions; thus, future research should further investigate whether emotional reactivity to stressors would exhibit as a curvilinear risk factor for predicting adolescent psychopathology. Finally, we considered the relationship between parent-adolescent conflict and adolescent emotional reactivity and psychopathology; however, recent studies have supported the bidirectional links between family systems in which adolescent characteristics would also influence the quality of parent-adolescent relationships (Chiang & Bai, Reference Chiang and Bai2022; Cortes-Garcia et al., Reference Cortes-Garcia, Wichstrøm, Viddal and Senra2019). Assessing the nature of bidirectionality in sensitization hypothesis is an important direction for future research.
In conclusion, this study makes theoretical and empirical contributions to the extant literature. Current theoretical frameworks of sensitization hypothesis primarily focus on emotional reactivity to interparental conflict. In this study, we provided novel evidence of how conflicts with parents increase adolescents’ NE reactivity to peer conflict, which then predicted increase in depression and anxiety symptoms. The findings also suggest that depressive symptoms are particularly related to PE reactivity to peer conflict, which is often neglected in studies of emotional reactivity or sensitization hypothesis. Moreover, this is the first known study to examine the unique effects of PE and NE reactivity to peer conflict on depression and anxiety symptoms, thus underscoring the importance of specifying both positive and negative experiences of emotions as predictors of developmental outcomes. Regarding translational contributions, the findings highlight the increased negative emotions in response to peer conflict is a target for preventive interventions that aim to reduce the adverse impacts of parent-adolescent conflict.
Acknowledgements
None.
Funding statement
Shou-Chun Chiang was supported by the Prevention and Methodology Training Program (T32 DA017629; MPIs: J. Maggs & S. Lanza) with funding from the National Institute on Drug Abuse. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute on Drug Abuse or the National Institutes of Health. Data collection was supported by the Taiwan Ministry of Science and Technology (MOST 110-2410-H-004 -109 -). MOST had no role in the study design, collection, analysis, or interpretation of the data, writing the manuscript, or the decision to submit the paper for publication.
Competing interests
None.
Ethical standards
The study was approved and followed by the Institutional Review Board of the National Chengchi University (NCCU-REC-202105-I038). The study is in accordance with the Declaration of Helsinki.
Consent
Informed consent was obtained from all individual participants included in the study.
Open access
